aheadCRM - Blog , Blog

Voice AI and Agentic Commerce: Your Personal Agent Rents Its Home From the Other Side of the Force

Thu, 03 Sep 2026 13:06:28 -0400

Every wave of customer-facing technology arrives with the same promise: this time the balance of power shifts toward the customer. As we discussed in our recent CRMKonvo with Dan Miller, founder and analyst emeritus of Opus Research, voice AI is making that promise again, and this time both sides of the counter are armed. The more interesting questions are where the customer's weapon is kept, who pays its rent, and who gets to whisper to it while it works.

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

https://youtube.com/live/2IQVkY2IVgU

Else, be my guest and continue to read.

Or do both …

Automated Voice Was Never Built for You

Dan has watched this market almost from its beginning, and he is blunt about its origin. Automated voice entered the contact center "almost entirely for cost savings". Everything after that was repair work: better speech recognition, task-specific tuning for finance and travel, and a long campaign to make a system optimized for cost do a passable impression of service.

The same instinct shows up in voice analytics. Once recordings could be mined, the first trigger enterprises asked for was a competitor's name, then a churn score. The retention arithmetic was well known; it’s roughly six times cheaper to keep a customer than to win one, and the wisdom that followed "got distilled into a few rules". Detect the person about to leave, offer them something.

In blunt words: the industry learned to close the door, not to fix the room. Nothing about generative AI repeals that. It runs the same rules faster, on more data, with better grammar.

The Same Data, Two Completely Different Businesses

Proverbially, technology is neutral and the intent behind the listening is not. Reading and analyzing a customer's conversations to make the customer the means to get revenue and reading them to make the customer successful with revenue as the consequence, are two mindsets that produce very different decisions from identical data.

Dan looks for the win-win and argues the crossover point is real: "successful firms do a better job of not exploiting their customers", and you cannot save your way into profitability. In a contestable market he is right. The trouble is how many markets are not. Starlink raises prices for the stated reason that it can, retires a plan and sells the stranded customers a dearer one. Banks make you dependent first and negotiate second; try collecting your wages in a brown envelope this year. Volkswagen shipped an ID.7 with a single set of rear window switches and a toggle to choose which window you are operating, because somebody found a few cents in the parts list.

None of that is a data problem, and no conversational layer fixes it. There are two ways to make a profit: by making me successful, or by making sure I make you successful. They are not mutually exclusive, but the slider between them is set by the executive, not by the model.

Your Personal Agent Has a Landlord

Dan's constructive answer is quite interesting. Build your own agent: one that knows your preferences, your payment methods, your preferred vendors, and goes into the market with your horsepower rather than the vendor's. Assume everyone else is a wolf. "I want this thing that is all mine and I can do battle on a better than even playing field."

He named both problems himself. The first is assembly. The people extracting real value from these tools are already technologists; everyone else gets a chat window. Do I buy the Mac with Apple silicon so I can run my own model? We have had the conceptual framework for a voice agent on a smartphone since Siri in 2009: seventeen years of the same slide, still no delivery.

The second problem is structural. Draft regulation in the United States already assumes a trusted, custodial user agent hosted on somebody's infrastructure: Google, Apple, your card issuer. Dan sees the consequence clearly: "there would be room for my agent to be influenced by these entities that don't have my best interest at heart." Preferential placement, promotional dollars, ad support, and then the slow enshittification Cory Doctorow named.

This is not speculation, it is a rerun. Social media was sold on exactly this rebalancing. The consumer held control right up until vendors understood what the tools were worth and bought the position back, because they had the funds. An agent you do not host is an agent you do not own.

If Agents Talk to Agents, Voice Shrinks to the Handoff

Here is my question for the voice AI market. If agent-to-agent commerce becomes the mainstream, why would two digital agents converse in a human language at all? Voice then survives only at the two edges: where I express intent, and where the result is handed back to me.

Dan broadly agrees. Ninety percent can be machine to machine and invisible; the remainder is rendered as speech, because it is the most natural interface for a human. Which is awkward, because the edges are exactly where voice has always broken. Ralf raised the automatic barista a Swiss food company built eight years ago: superb coffee, a microphone and speaker on the front, no chance of surviving a vandalised station concourse. Demo quality is not operational quality. Dialects, background noise and the oldest failure of all, which Dan still states best: "The last thing you want is going around and around because it didn't understand you."

The evidence on convenience is decades old and nobody likes it. Amtrak ran one of the first automated phone booking systems, and users programmed the touch tones into their speed dials. The fastest route through the voice system was not talking. Meanwhile customers consistently choose to sit on hold for a human rather than deal with the automation. That is a revealed preference, measured over thirty years, and it has barely moved. The reason is simple: The systems are designed about the company’s wants, and not the customers’.

Dan's engineering answer is the unglamorous one: list the ways the thing will fail, find the three to six modes that cause ninety percent of the failures, and fix those. It is the right answer. It is also not what the current round of voice demos is tuned for.

Intent Fulfilment Is the Only Meaningful Metric That Can Be Used By the Vendor

Dan's ice maker stopped working. He photographed the model plate, asked an LLM, got a wrong answer, sent more pictures, and eventually landed on unplugging the unit for thirty seconds. It worked. No ticket, no 800 number, no technician, no survey. Service happened and the manufacturer has no record that it did. Existing service delivery infrastructure, in Dan's words the people, the systems and the documentation, is what LLMs will eat.

Which is why the argument in Mitch Lieberman's forthcoming book, The Conversation is the Record, matters more than the agent hype around it. Put the conversation, with the context wrapped around it, into the system of record across every channel, and you can finally ask the one question worth asking. Dan's lightning-round metric was exactly that: intent fulfilment. Commitment made, commitment kept.

One can argue that a fulfilled commitment is not a finished job, because commitment doesn’t necessarily cover the intent. It still beats everything currently on the dashboard. Handle time, containment, deflection rate and NPS all measure how the enterprise felt about the interaction. Intent fulfilment measures whether the customer got what they came for, and it produces an auditable list of promises the company did not keep. That is precisely why no vendor is rushing to sell you the scoreboard.

Pragmatic Playbook for Enterprise CX Buyers

Asked who gains more influence over the next three years, Dan was honest: in his world, customers; in the real world, a standoff. Standoffs are decided by preparation.

For enterprise technology leaders evaluating Voice AI and agentic platforms, the market is currently a minefield of over-promised capabilities and hidden architectural debt. To avoid making a costly mistake, buyers must ground their strategy in three core operational rules:

Demand Intent Fulfillment Metrics, Not Deflection Rates. Stop measuring your CX success by how many calls your Voice AI prevents from reaching your contact center. Deflection is a cost-center metric that frequently masks customer frustration. Instead, instrument your systems to track commitment resolution: did the AI correctly capture the user's intent, trigger the appropriate backend API, and resolve the issue end-to-end? If your AI cannot execute real-world workflows inside your ERP and CRM systems, do not deploy it to customer-facing channels.

Decouple Conversational Data from Proprietary Vendors. Do not allow your Voice AI or contact-center-as-a-service (CCaaS) vendor to lock your interaction data inside their black box. Implement vendor-neutral conversational data standards such as vCons. Containerizing your audio, text, and metadata ensures that you retain full ownership of your customer interaction history. This allows you to audit AI performance objectively, feed pristine context into your RAG pipelines, and switch underlying LLM providers as better models emerge without losing your historical memory.

Establish Frictionless Human Escalation with Full Context. AI agents must never become digital dead ends. Design your conversational architecture so that the moment an agent detects intent ambiguity, user frustration, or an out-of-bounds request, the interaction transfers immediately to a human agent. Crucially, the human representative must receive the complete (Vcon container) transcript and real-time intent summary instantly, eliminating the infuriating customer experience of having to repeat information to a human that was already given to a bot.

Listening is cheap. Keeping your promise is the product.

The Inference Paradox: Tokens Got 1,000x Cheaper and Your AI Bill Went Up

Thu, 27 Aug 2026 14:24:38 -0400

The Inference Paradox:

Tokens Got 1,000x Cheaper and Your AI Bill Went Up

Somewhere in your organization there is a slide claiming that agentic CX is about to get cheap. I bet, there is. It has a line heading down and to the right, it cites the collapse in token prices, and it is not lying about the collapse. Token prices really have fallen off a cliff.

Still, the slide is wrong,

Why?

Because what you are buying is not tokens. It is workflows, and workflows have learned to consume tokens faster than tokens get cheaper. That is the inference paradox that Gartner Group talks about: the unit price falls, the invoice climbs, and neither number is a mistake.

The Price Collapse Happened Somewhere Else

Start with the part the vendors get right. Compiled inference-cost data from a16z, Epoch AI and Stanford's AI Index puts GPT-3-equivalent quality at roughly $60 per million tokens in late 2021 and about $0.06 by late 2024, a thousandfold drop, with price-performance improving at a median 50x per year and closer to 200x per year since the start of 2024. Goldman Sachs has semiconductor suppliers delivering 60 to 70 percent annual reductions in cost per token. Nobody can, nor does, dispute the direction of travel.

Now look at what the frontier costs today. Anthropic's published price list (as of August 27, 2026) puts Claude Opus 5 at $5 per million input tokens and $25 per million output tokens, and its largest models at $10 and $50. Those are the models your agentic workflow escalates to when the cheap one fails, and they are in a similar range as the GPT-3 launch pricing.

So the collapse is real, just that it happened at the commodity end. The price of last year's intelligence fell through the floor. The price of this year's did not.

And Agentic architectures are designed, to use this year's.

Token Inflation Is Measured, Not Alleged

This is where the research has caught up with the invoices, and the numbers are worse than most buyers assume.

Fu and colleagues gave the effect a name in Not All Tokens Are Equal: token inflation, the gap between advertised per-token pricing and what a workflow actually consumes once it retries what it got wrong. They measure inflation as high as 4.25x on multi-hop question answering, and they show that FrugalGPT, one of the standard cost-aware routers, underestimates true expense by more than 2x on hard tasks. This mechanism is what both drives the cost and nobody prices in: a failed reasoning chain is not only a wasted call, it is a call that gets re-sent in full, with history attached, to a more expensive model.

The infrastructure picture is even worse. Kim and colleagues measured what agents do to a serving stack rather than to a budget line in The Cost of Dynamic Reasoning. Tool-augmented agents make roughly 9.2 times as many model calls as a chain-of-thought baseline. A tree-search agent averages 71 calls per single request. Input sequences run three to four times longer because the interaction history accumulates. GPU memory per request rises three to five times, and the GPUs then sit idle 54.5 percent of the time waiting on tool calls. Energy per query rises 62 to 137 times over single-turn inference.

Gartner's 5-to-30x multiplier for agentic queries, which read like an analyst hedging, turns out to be the conservative end of the range.

This is the paradox in four short sentences:

The price per token fell by three orders of magnitude. The tokens consumed per useful outcome rose by one to two. Usage went up, way up. Compound those and you get a real invoice.

You Are Also Paying for Tokens That Make the Answer Worse

There is an additional line item nobody budgets for, and it is one a CX buyer should find most uncomfortable.

Zhou and colleagues studied what happens when you keep spending on reasoning in When More Thinking Hurts. Marginal utility on their test set drops from +1.8 percent per 500 tokens in the 2,000 to 4,000 range, to +0.1 percent between 8,000 and 12,000, and turns negative beyond that. Past roughly 7,000 tokens the model flips more previously correct answers to wrong than wrong answers to right. Their 32B model peaks at 55.8 percent accuracy at 12,000 tokens and falls back to 54.9 percent at 16,000. Easy problems start overthinking at around 1,500 tokens.

Read that again with a meter running. There is a point in every reasoning budget past which you are paying more money to get a worse answer, and it arrives earliest on the simple tickets that make up the bulk of your service volume.

The Efficient Agents work draws the same conclusion from the other direction: best-of-N test-time scaling buys marginal accuracy at disproportionate cost, and a leaner design retained 96.7 percent of the accuracy at $0.228 per problem solved against the $0.398 of richer systems. More compute is not a strategy. It is a default setting.

The CX Meter Hides the Multiplier

None of this would matter much if your contract passed the cost through legibly. It does not.

CX Today's buyer guide catalogues six live pricing models: per seat with tokens bundled, per channel, per component, credits, per action, and per resolution. Genesys runs $75 to $240 per user per month. Amazon Connect meters $0.038 a voice minute and $0.010 a chat message. Salesforce sells Flex Credits at $500 per 100,000, roughly ten cents an action, having launched Agentforce at $2 a conversation and repriced once that unit stopped fitting the work.

Zendesk's pricing page is different. It explains the outcome model clearly, that you pay only for requests resolved without escalation to a human, and it does not publish a rate. The headline unit of the most buyer-friendly-sounding pricing model in customer service sits behind a sales call and a tough negotiation.

Each model hides the multiplier at a different place; none of them is denominated in anything a CX leader actually manages. Per-message looks cheap until the agent generates repeat contacts and you are billed twice for failing once. Per-resolution looks aligned until you notice it pays the vendor not to escalate. Per-action asks you to model an entire workflow before you can forecast a quarter.

Jevons Was Not a Pessimist

None of this is an argument against spending. Cheaper units drive more consumption. That is Jevons. It is what happened to coal and steel and bandwidth, and it usually indicates a technology that works. Goldman expects token consumption to multiply 24 times, to 120 quadrillion tokens a month, between now and 2030. Spending more on inference in 2029 than you do today is not necessarily mismanagement.

The failure mode is narrower. It is not spend. It is unattributed spend. The organizations in trouble are not the ones with large inference bills; they are the ones that cannot say which workflow, which agent, or which resolved ticket a given bill belongs to, in other words, which outcome they pay for.

What to Implement Before You Sign

Look at preparing yourself using five measures, ordered by how quickly they pay back.

Instrument before you scale. Use per-agent and per-workflow token telemetry with alert thresholds, from the first pilot onward. One healthcare deployment ran from $12,000 to $68,000 a month over six weeks on a retrieval fault that went unnoticed for two of them. Cost you cannot attribute is cost you cannot defend, or avoid.

Cap the loops. Set hard retry ceilings with mandatory human escalation at the limit, and a fresh-escalation policy that discards a failed chain instead of forwarding it: Fu and colleagues found that passing failed reasoning to a stronger model cost up to 34.8 percentage points of accuracy. Uncontrolled retries are the single largest driver of runaway spend. You pay premium rates to make the answer worse.

Budget the thinking. Set per-task reasoning caps rather than letting a model run to its limit, and allocate them by difficulty rather than uniformly: Zhai and colleagues get up to 12.8 percent better accuracy on MATH at the same budget purely by varying compute per instance. Above the crossover point, you are buying degradation at full price.

Take the engineering discounts.Cache reads price at a tenth of standard input and pay for themselves fast. Route simple work to small models: one team took $40,000 a month down to $24,000 on routing discipline alone. Compact context, retrieve just in time, and stop shipping every tool schema into every call.

Put the meter in the contract. Get a defined billable unit in writing that names cached tokens, tool execution, failed calls and retries, not just input and output and audit rights to reconcile your own telemetry against the invoice. Demand a notice before any repricing or redefinition of the consumption model: current clause guidance suggests 120 days, and on a meter that can move mid-year I would ask for six months. Then measure value per thousand tokens against agreed outcomes rather than against volume. Organizations that do this report spending 60 to 70 percent less for equivalent output.

The tokens will keep getting cheaper. The bill will keep getting bigger. Only one of these two is under your control.

The Enterprise AI Intent Gap

Tue, 25 Aug 2026 18:39:46 -0400

Every hype cycle produces its own comfortable silence, and this one has a good one: almost everybody is doing AI, and almost nobody will say out loud whether it worked. In our recent CRMKonvo with Jon Reed, co-founder of diginomica, we spent an hour poking at that silence, with Ralf Korb doing the poking alongside me. Jon is one of the analysts who actually tests the thing before writing about it, which makes him tiresome company for vendors and excellent company for buyers. The conversation did not land on whether AI works. It landed somewhere considerably more uncomfortable: most enterprises cannot say what working would look like, and they started spending anyway.

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

https://youtube.com/live/YQkb-RPzvZs

Else, be my guest and continue to read.

Or do both …

Sixty Percent, And Nobody Is Blushing

Jon opened with numbers rather than opinion, a habit more of us should copy. McKinsey's recent work on AI measurement found that nearly eight in ten companies are using generative AI in some capacity, while around sixty percent report not seeing enterprise-wide EBIT impact from those programs. The gap between activity and impact is apparently not closing. Instead, it seems to be widening. A small group of over-performers is automating end-to-end workflows inside specific domains and getting results, and even they argue about what to measure and how to attribute the improvement.

Let that sink in for a moment. These are organizations that committed budget, headcount and executive credibility to a program, then discovered they had never agreed on a definition of success. Jon asked the obvious question: "Why would you undertake a project like this if you had no idea how you were going to measure the success of it?"

The answer is not stupidity.

It is fear.

There is "a profound fear of missing out or being left behind", and the people applying that pressure are usually the ones furthest from the technology. Executives and board members "have some of the most unrealistic ideas about AI in the entire organization". So the CIO is told to spend on something, anything, and the something arrives in the shape of a forward deployed engineer. Fine role, wrong instinct: a forward deployed engineer is a technologist. They do not know your business, and this was never an engineering exercise.

The Model Stopped Being the Product

This is the part of the story most vendor keynotes skip. Five or six years ago, scaling language models produced results that looked like emergent intelligence, and the valuations followed: if you can build truly cognitive systems, you can replace large parts of the workforce and justify almost any capital expenditure. What actually arrived is "a facsimile of intelligence". Then the scaling laws slowed, the training data ran thin, and investors got nervous.

What the labs did affects you more than the models themselves do. They wrapped the models in compound architectures with external verification, symbolic tools, deterministic cross-checks that refuse to let an agent post to the general ledger when the entry fails a rule. The industry sells this as context engineering and harness engineering. Strip the vocabulary away and the model has become the least interesting component in the stack. The architecture around it has become the product.

Jon's summary of the exercise is simple: "We're taking a tool that was not intended to be deterministic, and we're trying to see how far we can push that." That is a candid description of the state of the art, and it carries a warning no vendor slide will show you. Guardrails work in one direction only. "It's a lot easier to stop agents from doing something wrong than to know for sure that they did something right, because they don't understand what the right thing is." Blocking a bad output, better one too many than missing one, is engineering. Certifying a good one is still your problem.

Expertise Is Not a Commodity, and Nothing Is Learning

Two claims circulating on LinkedIn got taken apart, and both had it coming.

The first is that expertise has been commoditized.

It. Has. Not.

What has been commoditized is working-level knowledge across many domains, which the models absorbed during training. Working knowledge is not expertise. "It's only expertise that can identify the problems in the model output," Jon argued. That sentence should reorganize your hiring plans. If you believe the machine is the expert because it passed the bar exam, you will ship its mistakes at scale and file the result under productivity. Mathematics is an exception, because synthetic data works inside the closed confines of maths. Your industry is not maths.

The second claim is that the agent learns from your users.

It does not.

The language model is not learning while you talk to it. It is pre-trained, then trained, then frozen, and adjusting weights on the fly runs into catastrophic forgetting; Rich Sutton has a Turing Award and a working explanation of why. When a vendor says the system learns continuously, they mean that a knowledge graph or memory store are updated with your preferences. That is a storage mechanism disguised as learning. It is dangerous because it gives buyers the wrong idea of what is possible, and having wrong ideas about the possible is how budgets get burned.

The CSAT Trap: When Nothing Got Worse Counts as a Win

Now to CX, where the reasoning gets worse rather than better. We looked at companies that replaced level one service with AI assistants, reduced headcount, and reported the result as a win because their CSAT scores did not go down. Hold that up to the light. The stated ambition was to change nothing about how customers experience you while spending less on them.

"Fine, but you're not Amazon." Unless you are a behemoth or an airline, service is one of the few places where you can still out-compete companies that have more money than you. The question is not whether the bot held the line at nine in the evening. It is whether these tools let you run the best service in your industry, including at nine in the evening when your people have gone home.

Corporate intent decides that outcome. If the corporate desire is to solve issues, the technology gets designed to solve issues. If the desire is to deflect them, you have bought a deflection machine with better grammar. And the failure mode is almost never the answer itself; it is the escalation. Jon's own pharmacy routes him through a voice system that offers to help, asks him to describe the problem, then loops him back into the automation he was trying to escape: "If the automated system had answered my question, I wouldn't be asking to talk to the frigging pharmacist." Every enterprise reading this has built that loop somewhere.

Not every customer warrants the same treatment either, and pretending otherwise is not fairness, it is laziness. Your largest account should probably not be routed into the same voice system as everybody else.

Architecture Follows Intent

The best line of the hour was not Jon's own. He borrowed it from a diginomica colleague writing about the International Rescue Committee's AI operating model: architecture follows intent. Decide who you want to be, then build the thing that makes it possible. Most enterprises run that sequence backwards, buying architecture and hoping an intent turns up later.

Equifax came up as the counter-example, and the detail is the useful part. They credit their AI results not to a clever agent but to five years of cloud migration that left their data in a standard fabric, plus proprietary data the models have never seen. Nobody sensible will tell you to spend two years modernizing before touching AI. Jon did not, and neither will I. But the modernization track and the AI track run in parallel, and the sprinkle-sauce theory, the one where AI lets you skip the discipline, is "a LinkedIn feed fantasy land".

Pragmatic Playbook for Enterprise CX Buyers

Settle three things before the next AI proposal reaches your desk.

Build the evaluation suite before the program office. You have to have transparency over what your AI is doing. Define the business outcome, the baseline and the attribution method before the contract is signed, not after the pilot disappoints. Pick a problem meaningful enough to matter and contained enough that getting it wrong does not break the business. If nobody in the room can state success as a number, you are not ready to buy.

Put your pricing and your data in the contract. Any change to outcome-based or consumption-based pricing requires six months of notice so you can adjust. Moving off user-based licences to pay for tokens with no business result attached is not an advancement, it is a higher invoice. And when the vendor says their agent learns from your users, ask these two questions: how exactly does it learn from my users, and how do you protect that data? An update to the knowledge graph is not learning.

Design the escalation first, then hire someone to check the whole thing. Most customer anger at AI support is not about the answer, it is about being unable to get out. Build the route to a human before you build the deflection, and keep your most valuable accounts out of the automation entirely. Then consider the role Jon would add to the org chart: an AI ombudsperson whose job is to walk into departments, gut-check what is being built, and flag the vulnerabilities and the opportunities nobody else is positioned to see.

Architecture follows intent. Buy the intent first.

Gartner Group: Lawmaker, Judge and Executioner?

Fri, 21 Aug 2026 10:51:48 -0400

Gartner rewrote the CRM rules this year. It was probably right to. Buyers still need to read the fine print.

Gartner's 2026 Magic Quadrant for CRM Sales Platforms is likely the most consequential edition in years. But it isn't because of the vendor movements. It's because of the rule changes that caused these movements.

Let me start with the part that deserves credit. Gartner correctly saw the market shift and acted. The report was renamed from Sales Force Automation Platforms to CRM Sales Platforms, and the substance moved with the name. The old report was defined around records: leads, accounts, opportunities, pipeline, quotes, partner portals. The new one is defined around orchestration and composite AI, whether predictive, generative and agentic capabilities actually feed each other, and whether one can see, govern and correct what those systems do.

It is the right call. Anyone who has sat through a vendor demo in 2025 or 2026 knows the gap between "we have AI" and "our AI composes, gives results." Gartner's Trend 1 spells it out: most agentic capability today is "predefined graphs of large language model nodes, deterministic triggers, and text queries authored by administrators," and broadly reliable autonomous agentic selling is "more likely a post-2026 market development." That is a remarkable assessment from a firm whose clients would prefer to hear the opposite.

So: correct diagnosis, and a good response.

Here's the problem. In this market, Gartner writes the law, sits as judge, and carries out the sentence. And this year, the law changed substantially.

The scale of the rewrite

Put the 2024, 2025 and 2026 editions side by side and the change is unmistakable.

Four mandatory features were deleted. Collaboration, guided selling, partner relationship management and proposal/quote builder were all mandatory in 2024 and 2025. In 2026 they are gone. PRM wasn't just a feature; it was also an inclusion criterion. It is now worth nothing.

The entry criteria changed more than just a little. 2024 and 2025 asked for AI/ML features in three critical capabilities. 2026 asks for composite AI with at least two modalities in production, with at least two workflows demonstrating cross-modality operation where one modality's output informs or triggers another. A new "native baseline" clause was added: no third-party product may deliver core functions or the AI modalities used to qualify.

The coverage bar roughly doubled. Live implementations went from two of three use cases to four of five. Major releases required in twelve months went from two to three.

On top of this, six evaluation criteria were downgraded across two editions, with zero upgrades. Customer Experience fell from High to Medium. Marketing Strategy from Medium to Low. Business Model from Low to Not Rated. Then in 2026, Marketing Execution went to Not Rated, Sales Strategy to Low, Operations to Low. Every change moved in the same direction: away from commercial standing and go-to-market, toward demonstrated product. Which actually is a good thing.

But: six downgrades, no upgrades. That is not drift. That is a redefinition of what the market rewards, in Gartners opinion.

To be fair: the notice was published

Gartner did not spring this. It announced what will happen, not only once, but twice.

The 2025 edition carries a note to clients: the team has "chosen to place a heavy emphasis on AI capabilities," and "all write-ups, placements and scores in this Magic Quadrant and its companion Critical Capabilities reflect this new scoring approach." Then, in the same report, Gartner explained why Freshworks was dropped: the methodology "has become more product-centric — placing greater emphasis on vendor demonstrations, including but not limited to API payload demonstrations."

Freshworks was the proverbial canary bird. A vendor was removed in 2025 precisely because it could not survive a demo-centric methodology. That was a warning shot, fired a year before the titans got hit.

More than that, Gartner telegraphed the specific failures. Its cautions turned out to be a criteria roadmap.

Salesforce was cautioned in 2025 for "limitated AI sophistication and cohesion", saying that AI capabilities that were "disjointed, lacking cohesion between predictive AI and semantically driven recommendations." In 2026, composite AI became the entry criterion for the entire market. Salesforce closed the gap in one cycle and held Leader.

Microsoft read part of the memo. Gartner's 2025 caution was pointed: agentic demonstrations "highlighted agentic AI use cases outside of sales, such as the McKinsey & Company Onboarding Agent, raising concerns about Microsoft's internal AI agent playbook for sales." In 2026 that was fixed. But mobile has been a Microsoft caution for a while, and Gartner now calls mobile-first AI design "structural" and something that "cannot be easily retrofitted." The company still remained a leader.

HubSpot cleared the new bar. Its composite AI now hangs together, with conversation intelligence feeding next steps, prospecting and data agents working the same pipeline. This is precisely what the 2026 entry criterion demands. But Gartner told it in 2025 that guided selling relied on "static rule-based workflows not AI-driven recommendations," and the 2026 verdict on agent depth is barely softer: Breeze agents remain "constrained by manual prompt logic and narrow execution paths," with buyers advised not to expect "sophisticated autonomous orchestration, self-evolving agent behaviors or the ability to deploy extensive custom action libraries." It’s worth noting too that visualization and analytics was a HubSpot strength in 2025 and is a caution in 2026. Same product, higher bar. Still an upgrade from Niche Player to Challenger.

SAP did not read the memo. Its 2025 caution named "reliance on add-ons and integration... Microsoft Teams for conversation intelligence." In 2026 Gartner converted that sentence into an entry criterion, and SAP arrived with the identical dependency: conversation intelligence "relied on postcall Microsoft Teams transcript analysis." This earned SAP a downgrade from Challenger to Niche Player.

Oracle did not either. Its conversation-intelligence stitching was flagged as far back as 2024. Nine consecutive years in the Leaders quadrant ended over a gap named two editions earlier.

SugarAI got the loudest notice of them all. When Gartner announced its AI rescoring in 2025, exactly one vendor moved quadrant that year: SugarCRM, from Challenger to Niche Player. The reason was that administrators "cannot adjust model parameters, create custom prompt templates or choose data sources." Twelve months on, the platform "lacks a comprehensive framework for agentic orchestration and administrative oversight," with no native tools for "agent development, knowledge tuning, action-library configuration, composite AI, natural language analytics, or granular AI monitoring." The gap widened against criteria that now make it structural rather than cosmetic.

The vendors that moved up read the caution lists and shipped against it. That is the most useful thing in these three reports, and it is entirely actionable.

Where the three roles collide

Now the uncomfortable part.

When the lawmaker, the judge and the executioner are the same institution, a rule change doesn't just re-score vendors. It moves them, commercially, without anything about them changing.

Zoho's top-listed 2025 strength was its PRM portal. PRM stopped being scored. Zoho simultaneously closed a caution it had carried earlier: "basic AI-guided selling" and now earns credit for a "unified Zia experience". This is the exact cohesion SAP and Microsoft are still being cautioned on. It improved capabilities and moved from Visionary to Challenger.

HubSpot shed two cautions without doing a thing: guided selling and proposal/quote simply ceased to be criteria. Meanwhile high-velocity inside sales, its home turf, became one of five required use cases. Niche Player to Challenger, the largest jump in the report.

Oracle's mobile app was a documented strength in 2024 and again in 2025. In 2026 it is a caution. Oracle did not degrade its mobile app. The bar got lifted instead.

None of these are errors. It’s all justifiable. But collectively they mean that quadrant movement is a poor proxy for product movement – at least this year. In addition, vendors have no appeal, no external audit, and in many cases are also paying clients of the firm doing the judging. Gartner publishes an independence statement and takes it seriously. The structural tension still is there.

There is also the evidence standard itself. The 2026 report grounds nearly every caution in the phrase "Gartner-observed demonstrations." That is more transparent than the old approach, and it is also more cautious: "did not demonstrate " is not the same as "cannot do ." I wouldn’t be surprised if vendors invested heavily in demo choreography for 2027, to degrade this signal as it becomes a primary one.

Breadth beats depth, and that's an editorial choice

One more thing deserves attention. Moving from two-of-three to four-of-five required sales motions, natively, rewards generalist breadth and penalizes specialist depth, independent of scale.

monday.com and Vtiger qualify. ServiceNow does not, never has. Yet Gartner's own trends section argues that context federation is the next architectural battle, and that the cross-application overlay wins. That validates ServiceNow’s orchestration-layer thesis, while its clearest exponent sits outside.

That is a legitimate scoping decision. This is a sales platform Magic Quadrant, not a revenue orchestration one. But buyers should not read absence as a capability verdict, and they should notice that the gate and the narrative are pulling in different directions.

What buyers should actually do

Four things.

And this applies throughout analyst reports, not only this one.

Compare editions, not dots

A vendor that moved may have shipped nothing. A vendor that held may have closed a serious gap. Read the 2025 and 2026 cautions side by side; the signal is in the delta.

Re-weight the deleted criteria yourself

If you sell through partners, PRM still matters to you even though it no longer matters to the MQ. Same for proposal and quote, collaboration and guided selling. Gartner's criteria are Gartner's; your requirements are yours. Them not being assessed merely means that they are not shiny enough.

Treat the cautions as a forward roadmap

Cautions have predicted the following year's criteria three cycles running. Ask your shortlist vendors directly what they are doing about theirs, especially where they become interesting to you.

Test on your own data

Gartner says this itself in Trend 4, and it is the single most valuable sentence in the report: buyers must determine whether "their own data model, permissions, integrations, governance practices, and commercial entitlements can support the same experience" shown in a demo.

Gartner got the market call right this. It changed the rules because the market changed, and it indicated it in advance. That deserves acknowledgment.

But a rules change of this magnitude, adjudicated by the same body that wrote it, on evidence only that body observed, is not a neutral measurement. It is a considered opinion, which is exactly what Gartner's own disclaimer says it is.

Read it that way, and it is likely useful. Read it as a scoreboard, and you will buy the wrong thing.

The Customer Journey Illusion: Stop Mapping and Start Enabling

Wed, 19 Aug 2026 09:35:03 -0400

Welcome to another reality check. The CRM industry loves a good fairy tale. The most persistent one is the mythical customer journey. We like to pretend that customers wake up, look at our perfectly designed product pathways, and cheerfully walk down the well-paved brick road. The reality is far less smooth. As we discussed in our recent CRMKonvo with Dr. Graham Hill, organizations are not managing customer journeys. They are rather managing their internal processes, dressing them up in customer-centric language. It is a comforting illusion for them. It also keeps the stock price stable and the consultants employed. But it does absolutely nothing for the actual customer.

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

Else, be my guest and continue to read.

Or do both …

The Myth of the Master Plan

Let us look at the numbers. Dr. Hill pointed out an interesting statistic from a major UK retail bank. When looking at the "manage my home finance" experience, only one-sixth of the interactions were on the actual mortgage pathway. Five-sixths of the interactions were customers trying to get help with estate agents, solicitors, or basic support. Yet the bank only cared about the mortgage pathway. They willfully ignored the vast majority of the customer's actual reality. Why? Because the bank is only interested in selling the mortgage. Everything else was viewed as an unnecessary cost instead of opportunity.

This is a fundamental flaw in modern CX strategies. We design for the happy path and happily ignore the real world. We are shocked when our highly polished onboarding process falls apart at the very first sign of customer complexity. We create rigid silos. Then we wonder why our churn rates skyrocket.

Maps vs. Reality

This brings us to the core problem of journey mapping. Mapping a journey assumes a static topographical reality. It assumes that if we just draw a line from point A to point B, the customer will obediently follow it. As a sailor, I can tell you that this is a dangerous way to navigate. You do not just draw a line on a chart and blindly sail forward. You look at the weather. You look at the currents, at the waves. You constantly adjust your route based on dynamically changing conditions.

Not only when sailing.

Customers plan their way based upon certain criteria, and they replan it every single day, every single moment even, because the circumstances change. Businesses, on the other hand, force customers onto a rigid track. When the customer inevitably encounters a storm, the business is nowhere to be found. In the ocean there is no fixed path. Why do we expect our customers to travel on rails? We provide a mapped path that is utterly disconnected from the underlying terrain.

The Requisite Variety Trap

Why do companies insist on this broken model? Dr. Hill points to Ashby’s Law of Requisite Variety. Providing a rigid pathway is cheap and manageable. If you only have one prescribed pathway to get a mortgage, you can manage the interactions and complexity. As soon as you enable customers to do what they actually need to get done, your complexity and costs increase exponentially. It becomes very expensive to try and provide everything for everybody.

So, businesses take the easy way out. They stick to the one pathway that works for them, even if it does not serve the customer well. They rely on the sad reality that the evil known to them is often better than the unknown evil. Moving to a new provider is a hassle. Customers stay where they are until it becomes so unbearable that they are forced to move. This is not loyalty. It is hostage-taking.

Dumb Automation and the AI Mirage

The push for automation is often driven by a desire to reduce costs. Companies follow the exact opposite of the Toyota Production System. Taichi Ohno taught that you make it easier for the worker, then faster for the worker, and only then cheaper for the company. Modern banks and telcos do the reverse. They implement what Hill calls "dumb automation" to make things cheaper for themselves. They make it faster for the company, but they make it infinitely harder for the customer.

When you force a customer to use a poorly designed app instead of talking to a human, you are not innovating. You are just offloading your operational friction onto the person paying you. This cost-cutting strategy is fundamentally flawed. When customers cannot get their problems solved through automated channels, they resort to other means. They call the support desk. They complain on social media. They switch providers. The cost of recovering from these failures is astronomical.

In blunt words: an automated dumb process stays a dumb process.

Now we enter the era of Artificial Intelligence – again. The hype is deafening. We are told that Generative AI and LLM technology will revolutionize customer service. We are told that chatbots (err, agents) connected to a RAG architecture will flawlessly guide customers through their issues. Let me be absolutely clear. If your underlying data architecture is a mess, an LLM will simply hallucinate solutions based on that mess. Using RAG to query a broken knowledge base will just give you highly confident, grammatically correct wrong answers. Best regards from Air Canada!

AI is not magic. It is a tool that accelerates whatever processes you have in place. If your process is designed to ignore five-sixths of the customer's reality, AI will just ignore them faster. True innovation in CX requires a solid architectural foundation. You need a unified data layer that provides a single, accurate view of the customer. You need integration across your entire technology stack. Your CRM must talk to your billing system. The billing system must talk to your support platform. Without this integration, your AI initiatives are doomed to fail.

Businesses must accept that customers do not care about their product pathways. They care about getting their jobs done. If businesses want to succeed, they must align their systems and processes to support those jobs. They must stop trying to control the journey and start trying to facilitate it. This is not a marketing problem. This is an operational and architectural challenge. It requires rigorous analysis, tough decisions, and a willingness to challenge the status quo. If you are not prepared to do that, you should probably just stick to writing press releases.

Strategic Recommendations for Enterprise AI Buyers

Here are the core learnings and recommendations for enterprise AI buyers who actually want to improve customer experience rather than just buying the latest shiny object. We are past the point of treating software like a magical incantation.

Focus on Architectural Integrity over Generative Hype

Do not be seduced by the promise of an LLM fixing your customer service overnight. Before you invest in any advanced AI, audit your data quality and integration points. If your CRM cannot communicate seamlessly with your CDP, your AI will fail. You must build a unified data architecture first. AI requires clean, structured data to function effectively. If you build on a cracked foundation, you will only automate your existing dysfunctions. Get your data house in order before inviting the AI guests into the living room.

Design for Exceptions and Keep the Human-on-the-Loop

Stop optimizing solely for the rigid product pathway. As said, the vast majority of customer interactions occur outside of your carefully mapped routes. Use technology to handle the predictable, routine transactions, but design your systems to seamlessly escalate complex issues to empowered human agents. Do not use automation to build walls between your company and your customers. Use it to provide a solution faster. A human-on-the-loop strategy is not a sign of failure; it is a recognition of reality. AI should augment your workforce. It should not isolate your customers.

Measure Customer Outcomes, Not Internal Efficiencies

Your metrics are probably lying to you. If you are only measuring handle time or deflection rates, you are incentivizing the wrong behaviors. You must measure whether the customer actually achieved their goal. Implement systems to track the entire lifecycle of an interaction, including the rework required when automation fails. Yes, that’s harder to measure. But, when you understand the true cost of bad automation, you will stop prioritizing short-term cost savings over long-term customer value. Enable the customer to achieve their goals, and the business results will follow naturally.

A customer is a consequence, not a means.

Creatio's AI CRM: Who Gets to Build the Next Agent?

Tue, 04 Aug 2026 18:21:19 -0400

Every AI CRM vendor selling into 2026 has an AI agent story by now. The differentiator is no longer whether agents exist, but who is allowed to build the next one, how long that takes, and what happens to the bill once it works. For decades, CRM has promised growth and mostly delivered data entry, decaying from a system of action into a system of record. The agentic shift changes that, and with it the questions buyers should ask. Let’s put Creatio's AI CRM to those questions, following a deal from lead to order to see how much orchestration ships out of the box and how much a revenue team must assemble. The findings are published in full in my report, AI CRM for Revenue Growth: Inside Creatio's AI-Native No-Code Platform.

The company behind the platform

Creatio is a privately held, AI CRM and no-code workflow automation company headquartered in Boston, founded in 2014 by Katherine Kostereva, who remains CEO. It ran as bpm'online until a 2019 rebranding, bootstrapped until its first institutional round in 2021. A $200 million round led by Sapphire Ventures in June 2024 lifted its valuation to $1.2 billion; total funding raised now stands at roughly $268 million, and it reported around 50 percent year-over-year revenue growth at the time.

Creatio employs around 1,000 people and sells through more than 500 implementation partners worldwide. The company’s partner program has held a 5-star rating in CRN's Partner Program Guide for eight consecutive years. Customers span more than 100 countries, among them AMD, Colgate-Palmolive, and MetLife, with millions of workflows launched daily.

One platform, two studios

The product serves marketing, sales, and service on a single unified data model. Creatio Studio sits on top, split into Business Studio for no-code applications and AI Studio for autonomous agents, both sharing one data, security, and governance model. An in-app AI Twin now lets end users build their own agents from an IT-approved library without leaving the CRM. What makes this an AI CRM rather than a CRM with AI attached is where the intelligence sits: Creatio combines predictive, generative, and agentic AI in a single Creatio.ai architecture, reachable by end users in natural language, instead of bolting a chatbot onto a system of record.

Two authoring patterns cover most agent use cases. Prompt agents are simple assistants defined by a natural-language instruction plus the tools and skills the agent is allowed to use. Workflow agents are multi-step processes built on the same drag-and-drop designer that powers the rest of the AI CRM. Both are built by the same business-side practitioner who already configures pipelines and dashboards. There is no separate developer queue, AI-specialist hiring profile, or code repository in the middle.

Creatio was named a Leader in Nucleus Research's November 2025 LCAP Technology Value Matrix, and it was the only Leader in Forrester's 2024 Wave for low-code platforms built for citizen developers. That recognition shows up in practice too: BSN Sports runs its entire deployment for 2,600 users with just three administrators, while Howdens rolled out to 7,000 users across more than 800 depots in twelve weeks. Nucleus has separately measured 61 percent faster lead response, 70 percent faster implementation, and 37 percent lower total cost of ownership against legacy systems. Industry editions — including an agentic banking Solution that provides the basis for a Banking Blueprint that covers onboarding, lending, and KYC/AML — extend the platform into regulated sectors.

The pricing bet

In 2026, Creatio introduced an Unlimited plan tied to its Unlimited Enterprise operating model. One subscription covers unlimited users, custom agents, applications, workflows, custom objects, and API calls as a single platform fee, with AI included rather than metered. Beneath it, credit-based consumption is the default and per-user licensing remain available; AI Studio and AI Studio Twin add no incremental license.

The test: five agents, one deal

To test the authoring claim directly, I looked at a five-agent scenario across a single deal's lifecycle, combining shipped Creatio.ai agents with customer-specific ones authored in AI Studio:

An ICP-fit agent and an engagement-fit agent jointly qualify inbound leads, built on Creatio's Account Research and Lead Scoring agent patterns, promoting a lead to sales-accepted once both clear their thresholds.
An opportunity-health agent layers S/M/L risk sizing on Creatio's native MEDDPICC scoring, reads the opportunity record and call transcripts, and gates stage advancement until the criteria are met.
A SPIN-style coaching agent proposes concrete next moves on a stalled deal but cannot act without rep approval.
A service-brief agent, built on the shipped Customer Support and Knowledge Base agents, compiles ticket history, sentiment, and invoice status into an on-demand pre-call summary.

All five are registered, monitored, and governed in Creatio's unified administration layer, with PII policy, approval gates, cost thresholds by agent and model, and audit logging applied uniformly, whether the agent shipped with the product or was authored in-house. Each customer sets the rigidity, from letting agents auto-transition stages to requiring a human at every gate.

How the competition does it

Most competing approaches to agent-building fall into one of three patterns:

an agent designer wired tightly to a fixed data model, as with Salesforce's Agentforce and ServiceNow's AI Agents;
a horizontal builder paired with a separate CRM, as with Microsoft's Copilot Studio and Dynamics 365; or
a pro-code toolkit that still needs the engineering capacity it was supposed to eliminate.

Each carries a trade-off: opinionated designers constrain any customer whose process diverges from the vendor's reference, horizontal builders mean stitching two governance models together, and pro-code toolkits demand the scarce engineers they promised to free up. Creatio's pitch is that collapsing the AI CRM, the data model, the process engine, and the AI authoring layer into one product, governed from one console, sidesteps all three.

Analysis

The architectural claim holds up on inspection: governance, authoring, and the AI CRM itself sit in one architecture rather than three, which is a structural condition most agentic CRM vendors talk about, but few actually deliver.

The Unlimited Enterprise pricing model is the more interesting bet, however. It shifts the conversation from seats to execution at a moment when every competing consumption model bends upward exactly as AI adoption succeeds. The caveat is that Creatio's own default is AI credit-based consumption, so the unlimited promise and the metered tier still need reconciling. Whether it holds as genuinely unlimited at scale is the open question I would flag for any multi-year commitment.

The weaker spots are predictable for a company this size. Brand recognition in the upper enterprise and the North American mid-market still trails the legacy CRM incumbents, and delivery runs through that partner network, where outcomes vary with partner maturity. Neither is disqualifying, but both belong in a buyer's due diligence.

The AI CRM category itself is still being defined, so the more durable test isn't feature count. It's whether this architecture and this commercial model survive contact with deployments larger than the reference customers cited above.

Want the full picture, including the complete five-agent scenario, the competitive comparison, and the SWOT? My full report is available for download here.

Fewer Graveyards, Please: Legacy, AI, and the Debt You Cannot See

Wed, 22 Jul 2026 14:32:39 -0400

Every so often a vendor conversation earns the word "useful," and this one flirts with it. On CRMKonvo #309, Pega's Matt Healy walked into a room of skeptics and, refreshingly, did not try to sell AI as pixie dust. He sold governance. Let me explain why that is the interesting part, and where the pitch still needs a second look.

First, the setup, because it is absurd. According to Matt, ninety-five percent of Fortune 500s still run a mainframe in some capacity. Roughly thirty thousand organizations are still on Lotus Notes. Healy mentions a government claims system running on hardware funded by a grant from the JFK administration, and a separate agency contracting retirees back out of retirement homes to keep the thing alive. This is the installed base that every "AI-native transformation" slide ignores.

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

https://youtube.com/live/06tAjaWtQMk

Else, be my guest and continue to read.

Or do both …

The graveyard problem, now with agents

Modernization has been on the CIO agenda since CIOs were invented. What changed, is that frontier AI does not mix with data and processes trapped inside sixty-year-old systems. But even worse: if AI lets you build faster, it also lets you fill Alan Trefler's famous "application graveyard" faster than ever. The Lotus Notes graveyard of the 2000s simply reopens as an agent graveyard, or a Claude graveyard; take your pick of tombstone.

Healy does not dodge this. He cites the now-familiar numbers on AI-generated code: roughly eight times more duplicated blocks, double the code churn, and about 1.7 times the vulnerabilities versus human-written code. Convenient for a platform vendor to quote? Absolutely. Wrong? Not so much. His conclusion is the sensible one: what works for a hacker building a toy on a weekend does not survive contact with regulated, mission-critical scale.

"Model, don't code," or low-code wearing an AI hat?

This is the core message. Do not let AI generate an application from the foundation up. Use AI to translate business requirements into a model of the business: the processes, the decisions, the rules. Then let a platform run that model consistently, so ninety percent of the application behaves the same across hundreds of apps and only ten percent is specialized.

Revolutionary? Not quite: this is model-driven development, wearing an AI hat. The new and genuinely valuable move is using AI early: analyzing legacy systems, gathering requirements, researching regulations, and doing it grounded on curated best practices rather than, in Healy's words, "who knows where out there on the internet." That is where the productivity actually lives.

Predictability is the whole ballgame

Businesses want predictability, and probabilistic models are by definition not predictable. Healy's stance is the adult-in-the-room part of the episode. Keep orchestration and governance deterministic; confine agents to tasks like summarization, document handling, content generation, and research; and check their work. Crucially, produce visual, explainable models rather than millions of lines of machine-translated code that no auditor can read. When only about ten percent of enterprises have compelling AI in production, and the two roadblocks are cost and risk, "explainable and deterministic" is not a nice-to-have. It is the entire permission slip.

The pricing tell

Healy spells it out. Token-based pricing measures how much thinking the model does, which is not tied to value at all, and this incentivizes vendors to make the model think more. Pega's, as well as other vendors’, counter is outcome-based pricing: pay per claim, or a percentage of your cost per claim, and burn as many tokens as you like. This is the most buyer-aligned idea in the whole conversation. One caveat is of course that outcome pricing is also a lock-in and margin play, and the real negotiation lives in how you define a "claim" or a “resolution” and what today's baseline cost supposedly is.

The three-month miracle

Then the headline claim. An insurance was quoted by a system integrator seven years and roughly $25 million for a lift-and-shift from COBOL to Java. Pega plus it’s AWS tooling, we are told, produced a working application in three months.

The lift-and-shift critique is dead right: machine-translating COBOL into Java that nobody can read just translates old debt to new debt in the cloud. But "a working application based on their mainframe in three months" is an extraordinary claim resting on a single vendor-chosen reference that also happened to speak at Pega's own event. Before anyone budgets around that number, define "working." Which of the seven systems? What stayed on the mainframe? Who maintained it the day after go-live? And note the honest piece Healy volunteers: mainframe-zero is a fantasy. High-volume, low-latency payment processing stays put. The play is to extract the customer-facing, longer-running workloads and leave the transactional core alone.

The debt you cannot see is architectural

Which brings us to the point that vendors would rather skip: Gartner's observation (behind paywall) that technical debt is increasingly becoming architectural debt. That distinction is important, because swapping COBOL for something modern is a code problem, while re-cutting your solution architecture is a business-continuity problem. You cannot simply stop the enterprise, rebuild the plumbing, and switch it back on. Healy's answer is reasonable if unglamorous: plan top-down for where the business must be in one, three, and five years, let AI do the bottom-up archaeology of what you actually have, and meet in the middle. His genuinely useful framing is that this rationalization work, the sort of thing that used to eat six to twelve months of enterprise-architecture effort, can now be compressed into roughly two weeks. That is the compression worth paying for. Just do not let the same speed refill the graveyard with agents nobody governs.

Before you sign anything: three notes for CX buyers

Buy the archaeology, scrutinize the miracle. The lowest-risk, highest-certainty value is in using AI for legacy analysis, requirements gathering, and regulation research. Fund exactly that. Treat "three months, one platform, done" stories as scope-defined case studies, not as your project plan. Ask what "working" means, what was left running on the mainframe, and who owns maintenance after the confetti settled. A demo is a promise; a reference is a data point; neither is your architecture.

Make explainability and determinism contractual, not aspirational. For regulated CX, the probabilistic parts belong at the task level, boxed in and checked, while the process and governance stay deterministic and auditable. Insist on visual, explainable models. If your vendor cannot show a regulator how a decision was reached, understand that you own that risk, not them. "The AI decided" is not a defense you want to offer an auditor.

Price for outcomes, own the baseline. Outcome pricing beats a token meter hands down, so push for it. Just remember the leverage sits in the definitions: what counts as a claim or resolution, what today's cost really is, and what happens when volumes move. Bring finance and a hard-nosed controller into the room early, not after the first invoice. And govern the citizen-developer and agent sprawl from day one; the alternative is watching your Lotus Notes graveyard reopen under new management.

An unusually grounded hour. Healy sells discipline, not magic, and that alone puts this ahead of most vendor pitches. Just read the three-month case study with your glasses on.

The Agentic AI Mirage: Why Your 'Personalized' Assistant is Working for the Vendor, Not You

Wed, 15 Jul 2026 13:00:00 -0400

The Ghost of Cluetrain

In 1999, the Cluetrain Manifesto famously declared that "markets are conversations." It was an inspiring, romantic notion that promised to democratize commerce, wresting power from faceless corporate monoliths and handing it back to a sovereign consumer. Fast forward to today, and that conversation has been thoroughly co-opted. What was supposed to be a bilateral dialogue has devolved into an automated, highly-optimized monologue. The emergence of agentic AI, which features autonomous software agents supposedly operating on our behalf, promises a return to that original democratic vision. But let us be honest: is this actually a revolutionary shift, or is it just another iteration of vendor-controlled slop designed to monetize our decisions before we even make them?

The dream of conversational commerce was simple: technology enables humans to speak to other humans at scale. Instead, the vendor community realized that humans are expensive, inconsistent, and prone to demanding fair treatment. The corporate response was to replace them with IVR systems, chatbots, and automated messaging. These tools were never designed to foster actual conversations; they were designed to create efficient deflection barriers. Now, we are told that generative AI and agentic systems will change all this by acting as our personal proxies. But will it come true?

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

https://youtube.com/live/yBQ4y-VzZtc

Else, be my guest and continue to read.

Or do both …

The Illusion of Agentic Agency

During our recent CRMKonvo with Dan Miller, founder of Opus Research, we wrestled with this paradox. We have been apocaloptimists when it comes to conversational AI, marveling at the technology's ability to improve our lives while ignoring its potential as a tool for corporate surveillance. The simple truth is that the economic incentives of surveillance capitalism remain unchanged. When a vendor provides you with an "autonomous assistant" to help you shop, that assistant is not working for you; it is a digital Trojan horse. It is programmed to maximize the vendor's margins, steer you toward high-commission partners, and dynamically adjust prices based on your historical data. They call it serving you better; in reality, it is just more sophisticated extraction.

This is where the asymmetry of power becomes glaringly obvious. The consumer enters the arena with a simple objective: to find a quality product at a fair price. The vendor enters with predictive algorithms, historical CDPs, and agentic bots designed to extract the maximum possible lifetime value from that specific consumer. When these two forces meet, it is not a conversation; it is a “negotiation” where one party has access to the other's entire cognitive blueprint. If your personal shopping agent is hosted, managed, or trained by the same corporate infrastructure it is supposed to negotiate against, your agent is effectively a double agent.

The Guardrail Paradox and the Friction of Safety

One of the most fascinating aspects of our discussion centered on the concept of guardrails. In theory, guardrails are designed to protect users, prevent systemic bias, and ensure compliance. In practice, they are a friction point. If you are a malicious actor, or a vendor looking to maximize short-term profit, you do not want guardrails; building and maintaining them requires computational and human effort. Consequently, the path of least resistance is to deploy systems with minimal oversight and dealing with the possible fallout later. When they put restrictions in place, they often reduce legitimate user choices instead of protecting the user.

This creates a bizarre scenario where the consumer is locked in a digital playpen, restricted by strict guardrails on what their agent can ask or do, while the vendor's algorithms roam free in the wild west of data exploitation. The guardrail paradox is that by trying to make AI safe, we often make it useless for the consumer while doing absolutely nothing to stop the vendors from using unbridled models to leverage market dynamics. It is an asymmetric conflict: the defensive side must comply with every rule, which are set by the offensive side.

The 'Trusted Agent' in a Corporate State

Senator Warner and others have proposed regulatory frameworks that would authorize approved entities, be they banks, credit card issuers, or the vendors themselves, to host "trusted user agents." This is a farce of epic proportions. How can anyone believe that a vendor-hosted agent will prioritize the consumer's interests? The Martech community has spent decades building systems to capture, analyze, and exploit user data. To expect these entities to host an objective, consumer-first agent is akin to asking the fox to protect the chicken coop.

Such proposals do not democratize AI; they institutionalize the power dynamic favoring the vendor, dressed up in the shiny new clothes of trusted agentic technology. The vendor-hosted agent will inevitably suffer from a conflict of interest. It will prioritize the products that yield the highest margin, mask competitive alternatives under the guise of "simplifying choice," and feed our preferences back into the corporate data lake. True consumer agency cannot exist within a closed corporate ecosystem. It requires independent, decentralized, and locally run models that answer to no one but the individual user.

Pay-to-Play Algorithms and the Opacity of LLMs

Let us look at a concrete example of how this plays out in the real world. Generative AI led to the discipline of GEO (generative engine optimization) to ensure being highlighted in search feeds and assistant recommendations. This is the reality of a black box. When you ask a modern LLM for a product recommendation, you have absolutely no way of verifying why it chose a particular vendor. There is no transparent ledger of recommendations. It is entirely possible that the recommendation you receive is the result of an agreement between the LLM provider and a corporation.

As long as these models remain opaque, any promise of objective personal assistance is a marketing myth. The algorithms are trained on data that is already heavily skewed by advertising dollars and SEO manipulation. Therefore, when an agentic bot uses it, it is recycling corporate propaganda, presenting it as unbiased advice. This is not artificial intelligence; it is automated salesmanship. To combat this, we need absolute transparency in how recommendation engines operate, including a public ledger of all corporate sponsorships and algorithmic biases that influence the output. A tall order.

The Scalability Farce of Manual Compliance

Even if we establish clear privacy guidelines, such as the right to be forgotten or standard opt-outs like in the GDPR, e.g., implemented using the IEEE My Terms standard, the enforcement mechanism is broken. If a consumer requests that their data be deleted or excluded from training sets, how do they verify compliance? They cannot. If you send a compliance request to a trillion-dollar tech company, that request likely lands on the desk of an understaffed compliance team using a manual process to scour databases, call transcripts, and unstructured chat histories. This does not scale. It is impossible for these enterprises to manually comply with millions of granular privacy requests.

The vendor's SOP will be to say they complied. Yet, once your data has been ingested into an LLM, it is practically impossible to "un-train" that model on your information. The data becomes an inseparable part of the algorithmic weights. Therefore, any regulatory framework that relies on retroactive compliance is a toothless tiger. We must shift the battleground from retroactive deletion to proactive, systemic prevention.

VCONs and the Architecture of True Data Sovereignty

If we want consumer agency, we must shift the paradigm. This is where technologies like Virtual Conversations (vCon) become critical. A VCON is a standardized, secure digital container that houses the transcript, audio, and metadata of a conversation. Crucially, instead of relying on a vendor's pinky-promise to respect our privacy, the data itself is encapsulated with its own governance rules. This is a step toward true data sovereignty, but it requires a massive cultural and technical shift.

Conclusion: Taking Back the Loop

The term "human-in-the-loop" is frequently used to describe safe AI integration. But as agentic AI evolves, we are moving toward a world where humans are removed from the loop, replaced by autonomous agents transacting with other autonomous agents. If we do not demand models that genuinely operate on our behalf, we will find ourselves shut out of our own decision-making processes. This is time to stop being passive consumers of AI convenience and start being active architects of our digital autonomy.

Pragmatic Playbook for Enterprise CX Buyers

Enterprise buyers are currently being bombarded with vendor pitches promising that agentic AI will magically solve their customer experience woes. If you are a buyer and concerned about ethical AI use, here is your survival guide to avoid making an expensive, possibly brand-damaging mistake:

Prioritize Architectural Integrity Over Hype: Do not be seduced by an agent's ability to generate natural-sounding excuses. Demand to see the integration map. If the agent cannot access your back-office CRM and ERP data securely and deterministically, it is not an agent; it is a glorified chatbot with a larger vocabulary.

Mandate Strict, Verifiable Data Boundaries: Ensure that your customers' data is never used to train a vendor's public LLM. If the vendor cannot guarantee and prove that your proprietary customer interactions are kept in a secure, isolated RAG environment, walk away. Your customer data is your competitive moat; do not give it away to train your competitor's next model.

Implement 'Agent-in-the-Loop' Safeguards: Autonomous agents are highly efficient at going sideways before they go south. Never deploy an agentic system in a customer-facing role without a deterministic routing mechanism that instantly escalates complex, emotional, or high-value interactions to a well-trained human agent, complete with full conversational context.

Insist on Standardized Metadata and vCon Support: Prepare for a future of decentralized data. Your architecture should support standard containers like vCons to ensure that as consumers demand greater control over their conversational data, your systems can comply programmatically rather than relying on manual, unscalable processes.

The Illusion of the AI Copilot: Why Your Legacy CRM Architecture Isn't Cutting It

Wed, 01 Jul 2026 09:25:49 -0400

For years, the enterprise software complex has sold us on a beautiful fairytale: the single source of truth. We were told that if we just poured enough capital into our CRM systems, and if we just badgered our front-line sales representatives enough to log every transactional interaction, absolute operational clarity would emerge. Now, the enterprise technology industry has found its next silver bullet: generative artificial intelligence. Every major software vendor is frantically bolting an AI copilot, a generic conversation summarizer, or an automated opportunity scoring engine onto their legacy applications. They promise that these shiny additions will magically transform messy, unlogged data into executive-grade operational insights. But let us be completely clear here: it is mostly marketing fluff designed to protect legacy vendor stock prices rather than solve foundational architectural bottlenecks.

The recent conversation on CRMKonvo with the co-founders of Brief Executive Intelligence cuts straight through this generative AI hype. Larry Augustin, Clint Oram, and Zac Spreckett are not starry-eyed AI tech evangelists; they are battle-hardened industry veterans who built SugarCRM and spent decades in the enterprise application trenches. Their core thesis is as brutal as it is interesting: CRM platforms were natively architected for front-line reps, not for the executives who actually manage the strategic direction of an organization. Bolting a generic large language model (LLM) onto a legacy database framework does not fix the fundamental structural deficiencies of that historical ledger. It merely allows corporate environments to generate summaries of incomplete information faster than ever before.

TL;DR

If you want to watch the full CRMKonvo, please go ahead here (optimized for smartphones) or here (optimized for tablets/computers).

https://youtube.com/live/nCs3ws2Kygk

Else, be my guest and continue to read.

Or do both …

Legacy Software Architecture and the Customer-Centric Trap

To understand why current corporate AI initiatives are stalling, we must evaluate the structural foundation of legacy software. Systems of record are fundamentally passive; they are organized around a specific external entity, which in the case of CRM software is the customer. They excel at capturing transactions retrospectively, functioning as a historical record of what your account teams managed to input between active sales calls. This framework works reasonably well for rear-view reporting, but it breaks down when an executive needs to make real-time asset allocations or make time-sensitive decisions.

When vendors attach an AI assistant to a standard CRM framework, that artificial intelligence remains severely restricted by the constraints of the underlying data model. The copilot can surface information about an active customer account—assuming your reps actually took the time to manually input it—but it maintains zero architectural visibility into what is occurring across the rest of the enterprise infrastructure. It possesses no conceptual awareness of internal product pivots hammered out in Slack, the roadmap modifications documented in engineering tools, or the critical budget parameters negotiated in email chains. The bolted-on AI layer is effectively blind to the context of the executive function itself; it optimizes for an isolated department while leaving the senior leadership team completely in the dark regarding cross-functional reality.

The Rampant Enterprise Epidemic of Decision Amnesia

This deep architectural gap manifests in what Clint Oram labels "decision amnesia." In modern corporate environments, organizational activity happens at a dizzying pace. Generative utilities have made it incredibly simple to mass-produce content, which in turn accelerates the sheer volume of daily communications an executive must filter. This structural acceleration creates a striking paradox: modern enterprises are communicating more than ever and are understanding less. Critical corporate decisions are finalized in frantic chat threads, impromptu video calls, and rapidly buried message strings.

Without an enterprise framework that treats these decisions, commitments, and strategic corporate goals as native, first-class data objects, these crucial operational elements simply evaporate into organizational noise. SAP calls this framework to avoid decision amnesia the company memory.

The long-term cost of this decision amnesia is staggering; entire corporate leadership teams spend hours re-litigating the exact same strategic issues they supposedly resolved weeks prior because no internal platform recorded the precise reasoning behind the original alignment, nor the precise agreement. Enterprises find themselves trapped in an operational loop, acting like corporate hamsters spinning a wheel fueled by an endless stream of AI-generated communication exhaust. Organizations become highly active, deeply exhausted, and yet structurally stagnant, moving nowhere.

Beyond Storing Artifacts: The Era of Continuous Understanding

This brings us to a critical architectural distinction: the difference between merely storing data artifacts and actively maintaining an ongoing understanding of work. Traditional software applications are exceptional at storing passive artifacts: a saved document file, a logged call note, or an archived email chain. But a massive collection of independent data artifacts does not equal true institutional knowledge. Without interpretation it remains mere data. Expecting a human executive to manually synthesize thousands of scattered communication artifacts into a coherent operational picture is a guaranteed recipe for immediate corporate burnout.

The alternative approach requires a technology architecture built around a persistent knowledge graph paired with targeted language models. Instead of waiting for a user to actively execute a search query in a blank text box, an executive-grade system must continuously monitor the operational tendrils of the enterprise infrastructure. It must automatically parse ongoing communications, extract underlying corporate commitments, map those vectors against explicit corporate goals, and maintain an ongoing semantic representation of corporate reality. This is not about building a better data indexing engine or expanding an LLM context window. It is about establishing a foundational technology layer that inherently understands how a corporate entity operates.

Proactive Intelligence Versus the Search Box Obsession

Most current corporate AI tools are completely reactive; they sit quietly in a side panel until an executive types a specific prompt into a search interface. But as any seasoned enterprise leader will tell you, an AI search query is only as good as the question you know to ask. If you are completely blind to a developing operational crisis or a slipping cross-functional dependency, you will never think of typing it into your AI copilot. Reactive software infrastructure keeps corporate leadership in a perpetual defensive posture, scrambling to address systemic vulnerabilities after they have already degraded the bottom line.

True executive-grade technology must pivot entirely toward proactive intelligence. The underlying software must understand your current corporate context – the strategic partners you are meeting with, the business accounts that are drifting, the internal commitments coming due – and actively surface relevant insights to you before you realize a gap exists. If you are preparing for an investor call or a board presentation, you should not be spending the prior evening frantically querying disparate data silos to compile a status brief. The platform should already have mapped the operational trajectory and prepared you for the discussion. This is the difference between a simple digital assistant and an enterprise intelligence layer that actively protects your focus and accelerates human execution.

Enterprise AI Buying Strategies: A Guide for CX Leaders

For corporate leaders navigating the chaotic market of enterprise AI, avoiding expensive mistakes requires rigorous architectural skepticism. Consider these three core recommendations.

First, audit the underlying data architecture beyond the copilot hype. When legacy vendors show AI that creates quick summaries, check where that information originates. If the AI merely queries a siloed database, it will never provide cross-functional context. Demand a unified knowledge graph that synthesizes disparate communication channels like email, calendar, and chat. Do not pay a premium for a thin conversational interface over bad data.

Second, prioritize proactive intelligence over reactive search utilities. A system that requires users to constantly query a prompt box is a system that fails them. Evaluate software based on its ability to surface insights autonomously using immediate context. Ask vendors how their platform alerts leadership to misaligned goals or slipping project timelines without requiring manual configuration. Eliminate the administrative burden of searching for information.

Third, insist on absolute data privacy and security at the user level. Executive context contains sensitive enterprise data like financial trajectories and board reports. A generic cloud solution that pools data or exposes it to manual vendor reviews is an unacceptable liability. Ensure a security model where data is encrypted individually with unique keys, preventing vendor access to corporate intelligence.

Usage-Based Pricing for Copilot Is Good for Microsoft's Investors. Read That Sentence Again.

Sat, 20 Jun 2026 14:52:55 -0400

TheStreet ran a piece this week arguing that, of Microsoft's two Copilot announcements, the shift to usage-based pricing matters more to investors than the DeepSeek flirtation. That read is correct. It is also the tell.

Here is what Microsoft actually did. Copilot Cowork, the agent that reaches across Microsoft 365 to run multi-step work on your data, is coming off the flat per-seat add-on and moving onto consumption billing the company calls "Copilot Credits." Charles Lamanna, who runs Copilot, told Axios the product could not be offered on an unlimited-use basis. The users he pointed to are the ones doing hundreds of tasks a week. He called them "way productive." And then he said the part vendors normally keep off the slide: their costs go very high.

So the most productive users are the expensive ones. Hold that thought, because the whole argument lives there.

What "good for investors" is really saying

A pricing model earns the label "good for investors" when three things are true. Revenue starts to track cost-to-serve. Revenue scales with consumption instead of sitting flat per seat. And the vendor stops eating the margin on its heaviest users. All three are true here. None of them is a statement about whether a customer got value.

That is the gap I want to sit in for a minute.

Usage-based pricing meters an input. Tokens, compute, credits, whatever the unit. The customer does not buy tokens because they want tokens. They want a finished report, a resolved ticket, a reconciled spreadsheet. The token count is the cost of producing the outcome, not the outcome. And the relationship between the two is loose at best.

TSIA put it plainly in its May analysis of AI pricing: usage does not equal value, and consumption models often fail to reflect actual business value. That is not a critic talking. That is a research firm whose audience is the vendors building these models.

Agents make the coupling worse, not better

A chatbot answers and stops. An agent keeps going. It reads files, calls tools, checks its own work, hits a wall, tries again. Each of those steps burns compute, and the steps are what get metered. Every retry, every verbose detour, every loop the agent runs to second-guess itself adds to the bill. The customer pays for all of it.

Now ask yourself the hard question. Does a workflow that took the agent three retries and a long chain of self-checks deliver more value than the same workflow done cleanly in one pass? Of course not. It delivers the same outcome and costs more. Under seat pricing, that inefficiency was the vendor's problem. Under usage pricing, it is line-itemed onto the buyer's invoice.

The billing platform Flexprice, which sells the plumbing for this, says it out loud to its own customers: retries, loops, and background jobs are friction, not value, and usage-based pricing only works when customers gain something real as the meter climbs. Their warning to vendors is the buyer's whole case.

The people who get punished are the people who bought in

We do not have to guess how this lands, because GitHub Copilot already ran the experiment. On June 1 it moved to token billing. The median user barely noticed. The pain landed on the top five to ten percent, and it landed hard: community projections of bills jumping ten to fifty times, one developer modeling a move from roughly $29 a month to nearly $750, another claiming $50 to $3,000. TechCrunch called it the end of Copilot's golden age.

Look at who those heavy users are. They are not abusers. They are the people who took the vendor's three-year advice to use the tool for everything, built agentic workflows around it, and made it part of how they work. The pricing change penalizes exactly the depth of adoption every vendor claims to want. And the old safety net, where running out of premium budget dropped you to a cheaper model so you could keep working, is gone. What is gone, too, is the cost ceiling.

Lamanna's "way productive" power user and GitHub's top-decile developer are the same person. The model charges most to the customer who is succeeding most. Reward and penalty have swapped places. The reward now goes to the vendor.

The structural problem, which is bigger than the bill

Now the second half, and this part is more important than any individual invoice.

Think about where accountability for value sits in each pricing model. With outcome-based pricing, the vendor gets paid when a result lands and not before. Fin's (formerly known as Intercom) Fin charges 99 cents per resolution, billed only when the customer confirms the AI actually solved the problem. Under that model, every failed attempt costs the vendor. So the vendor has a direct, financial reason to make the agent efficient, accurate, and sparing with compute. Their margin depends on it.

Usage pricing inverts that incentive. The vendor is paid for activity regardless of whether the activity worked. An agent that burns more tokens, retries more often, and reasons more verbosely produces more revenue, not less. I am not claiming Microsoft will deliberately bloat Cowork to pump credits. I am saying the financial pressure that used to push toward lean, effective agents has been switched off; and switched-off incentives have a tendency of showing up in the product eventually.

The demand side pulls in the same direction. There is a name for it now: tokenmaxxing, the workplace habit of treating AI usage as a proxy for productivity, where people get judged on how many token they burn rather than on what they shipped. Built In's writeup is blunt about it: the habit rewards visible activity, not results. So stack the three forces. Buyers under pressure to run up consumption as a status signal, a vendor that meters by consumption, and an agent that inflates consumption on its own. Everything drives the meter up. Nothing points it at the outcome.

That is the real cost of the model. It moves the vendor one step further from owning the question of whether you got value, and it hands that entire question to you. The vendor essentially plays Pontius Pilate. The buyer now runs FinOps for AI. You set the budget caps. You write the spending policies. You read the consumption dashboard. You type /cost to see what a task burned. Microsoft, to its credit, is shipping all of those controls, and they are better than the ones GitHub fumbled out the door. But notice what they are. They are tools for the customer to govern value. They are not the vendor guaranteeing it.

The honest counterargument

I would be doing the same vendor-spin thing I just criticized if I left out the other side.

Flat pricing for agentic tools is inherently unsustainable. The economics are upside down: the model subsidizes the heaviest five percent and overcharges the lightest fifty. Metered billing is the rational fix for that, and for a low-volume or experimental buyer it is a better deal than paying a fat seat fee to barely use the thing. Aligning price with cost-to-serve is a good thing. It is just a vendor virtue, not a customer one, and the trick to watch is anyone presenting the first as if it were the second.

Cheaper models do not solve the coupling problem. A fine-tuned DeepSeek on Azure lowers the unit price of the metered thing. It does not make the metered thing track value. You are paying less per token for a number that still has a loose relationship to your outcome. And who knows how many additional tokens a potentially inferior model burns.

Where this lands

On my orchestration battleground, Cowork is the M365 layer that coordinates work across your apps and your Graph. Pricing that orchestration by consumption reframes it from a capability you own into a utility you rent by the drink. That is a substantial shift in who carries the risk when an orchestrated workflow goes long, and it is the buyer.

So, on the null hypothesis. Is usage-based pricing good for customers? Largely no, and for the reasons the question assumed. The metered unit is loosely coupled to value, agents widen that gap rather than closing it, the model bills the most engaged users the most, and it relocates the entire burden of value accountability from the vendor onto the buyer. Good for investors and good for customers are not in alignment here. On this one, they partly trade off.

Three things to do if you are buying.

Model your power users, not your average. The average user will not break your budget. The fifteen people who actually adopted the thing will, and they may very well be the ones delivering your return.

Make the vendor define the unit before you sign. If a task can cost anywhere from a few credits to a few hundred depending on how many times the agent talks to itself, that is not a price, it is a range. And you'll end up at the upper end, trust me. Ask vendors to commit to a per-outcome cost and watch how fast the conversation gets vague.

Push for outcome terms on anything that has a definable outcome. Resolution, completion, ticket closed. If the vendor will only price the effort and not the result, they are telling you something about how confident they are in the result.

The interesting question is not whether usage pricing is here. It is. The question is whether buyers will accept a model where the vendor is paid the same whether the agent nails it on the first try or flails through ten, or whether the market pushes back toward paying for outcomes the way Fin does. Or Zendesk. Or Hubspot. Or others. I do not know which way that goes. But the vendor whose margin improves when its agent works harder is not, structurally, the vendor most motivated to make the agent work better.