Many value teams are using AI to build business cases faster.
The leaders are using it to build something far more valuable: an intelligence layer that makes every seller and customer success rep sound like their best value engineer.
That was the clear signal from our Q4 2026 Business Value Collective Jam Session, where more than 100 value practitioners signed up to compare notes on AI, value selling and the build versus buy dilemma. No pitches, no spin. Just hard-won lessons from people running value programs at scale.
Here are the Six Value AI Trends that stood out, and what they mean for your value practice.
1. Start in the middle, and lead with the story
The most effective AI rollouts do not start at the top of the funnel. They start where a champion and an economic buyer are already engaged, and the seller needs to hand over a narrative that sells when they are not in the room.
One enterprise value leader put it bluntly: a great business case is 80% story and 20% math, and the math comes last. Numbers without a narrative do not land. Start with why you are even talking, then prove the juice is worth the squeeze.
The foundation is a value strategy map: the customer’s priorities (grow, run efficiently, reduce risk), their named initiatives from the 10-K and investor materials, the use cases you map to them, and the KPIs with a target and a timeline. Without the timeline, there is no urgency.
Win adoption in the middle, then extend outward to account planning upfront and value realization downstream.
2. Vibe coding is cheap. Governance is where it can get expensive.
AI quantification has become far more predictable. You can vibe code a credible value calculator in an afternoon. That is exactly the trap: most teams are pricing the build and ignoring the enterprise cost of owning it.
The hidden costs surfaced again and again:
- Consistency: homegrown tools ship ROI figures that can swing from crazy high to crazy low, with guardrails that often aren’t in place.
- Change control: who updates the model, how changes are versioned, and what happens when the builder leaves.
- Security and IT load: your value tool joins the queue behind every other AI project when something breaks.
- Co-creation: you will not share your personal AI project with a customer, so the business case stays thrown over the wall.
Where is the crossover? My working rule: a single value engineer supporting 10 to 20 sellers can vibe code effectively, as long as intelligence is captured and the logic is governed.
Beyond that, or at one or more cases per seller per week, you are rebuilding a platform feature by feature. The real question is not build versus buy. It is who owns the value workflow.
3. Intelligence beats tools
The most provocative idea of the session: value models and calculators are the engine, but the lasting edge comes from a brain that learns from every use.
Every business case, discovery call and transcript is a data point. Aggregated, they become your benchmarks: what resonates by industry, what improvement percentages are realistic, and what good looks like when a customer asks 6, 12 or 18 months later, “How am I doing against what you promised?”
This is where homegrown builds are weakest. They answer one seller’s question today but cannot tell a CRO or CFO what is working across the team. One practitioner described the goal as a continuously learning system that prompts sellers in the moment: you are missing the budget owner, you have not validated this objective, here is the hypothesis to pressure test next.
The UI matters less than ever. With MCP, the value repository can feed research, Slack, CRM and even presentations directly. The asset is the governed intelligence underneath.
4. Show up with a point of view
Great business cases come from great discovery. But the group was clear that “What keeps you up at night?” is a credibility killer.
The better path is outside-in: understand how the customer makes money, do the homework, and arrive with a hypothesis. People correct a point of view far more readily than they answer an empty-handed questionnaire. Even a partial value hypothesis, with the missing inputs clearly flagged, gives buyers a reason to engage in the next round of discovery.
AI makes this scalable. Structure the industry and role point of view once, then let AI super-personalize it with account context. That is how you earn the right to the conversation when AI-generated outreach is flooding every inbox.
5. Speed is not the value proposition
“Business cases in minutes, not days” is everywhere. One participant challenged it with a sharp analogy: you would not pick a heart surgeon because they finish in 30 minutes instead of four hours.
Speed does matter in one place: disqualification. A fast, early value hypothesis that does not land tells you to close-lose the deal before it drains the forecast. Beyond that, better discovery may actually lengthen a stage. Go slow to go fast.
So stop selling AI value on hours saved. Leaders are measuring what matters to the CRO and CFO:
- Win rate and forecast reliability
- Deal size and reduced discounting
- Renewal and expansion
- Cost to generate a dollar of revenue
6. Value is becoming a capability, not a function
Value started in product marketing, then moved into sales and presales. Now a new pattern is emerging: at least two enterprise teams on the call have paired business value with enablement, under banners like go-to-market excellence.
The logic is compelling. Most companies do bespoke value engineering on their biggest deals very well. Few scale it to every seller and every customer success interaction. Pairing value with enablement shifts the mandate from producing documents to embedding value selling into just-in-time learning and the daily workflow.
As one leader framed it, there cannot be product enablement over here and value selling over there. It has to be one integrated capability, delivered when the seller needs it.
The Bottom Line
The teams pulling ahead are not asking “build or buy?” They are asking what intelligence they need to own, and how to put it in every seller’s workflow. Five moves to make this quarter:
- Codify the story first. Build or refresh your value strategy map before automating anything.
- Start in the middle. Equip champions with a narrative they can carry internally, then extend outward.
- Set your crossover rule. Decide when a vibe-coded tool graduates to a governed platform, before governance gaps force the decision.
- Capture everything. Every case, input and outcome should feed benchmarks you can defend to a CFO.
- Measure outcomes, not hours. Report win rate, deal size and net retention, not minutes saved.
A big thank you to Mike Samuels, Global Head of Business Value at Datadog prior(@ Pendo, Smartsheet, ServiceNow, Oracle, SAP, KPMG) for getting the jam started, and sharing his amazing AI experience and insights with the BVC.
Book a one-on-one to pressure test your AI-powered Value-Led Growth use cases and strategy: https://calendly.com/tpisello-gd/one-on-one