Does AI make your leads convert better and faster?

Of all the questions B2B revenue teams are asking about AI right now, this is the one that decides the budget. If AI-influenced buyers convert at a higher rate and close faster than the rest of your pipeline, AI is a tailwind, and the work is figuring out which content and positioning is creating that tailwind so you can do more of it. If AI-influenced buyers convert at a lower rate or stall longer, AI is a detractor, and the work is figuring out which specific framings AI is propagating about you so you can counter them.

The mistake most teams make is trying to answer this question without first instrumenting the cohort. Pipeline tells you who closed; it doesn't tell you who consulted AI on their way in. Without a way to tag the AI-affected cohort, every other attribution signal you collect is observation without controls.

This guide is the deep dive on the third of the four attribution questions we use at Unusual to frame the full picture. It pairs with the Complete Guide to AI Attribution, which lays out the broader framework. Read that first if you want the context.

The headline signal: the AI-affected cohort test

The single move that opens up every other answer in this guide is tagging the AI-affected cohort. Once you can compare AI-influenced leads against the rest of your pipeline, the conversion question becomes empirical instead of anecdotal.

The setup is light. Add a second question to your intake: "Did you consult an AI model when researching us?" This sits alongside "How did you hear about us?" and produces a flag on every lead. For self-serve and mid-market segments, putting both questions on the intake form usually works. Enterprise buyers might be a little embarrassed to admit they used AI to evaluate you, so hold the AI question until you've built some trust. A discovery call or a second touchpoint works better than the form.

Tag the leads who say yes and compare that cohort's pipeline performance against the rest on four dimensions:

• Conversion rate by stage. Where does the AI-affected cohort fall out of the funnel relative to baseline?

• Win rate. When the AI-affected cohort gets to proposal, do they close at the same rate?

• Average contract value. When they close, do they buy more or less than the baseline?

• Cycle time. How long does the AI-affected cohort take to close compared to the rest?

Higher conversion in the AI-affected cohort means AI is qualifying buyers on your behalf. Lower conversion means AI is disqualifying them in conversations you will never see.

The shape of the cohort's funnel tells you more than the headline does. If AI-affected leads book demos at the same rate but close twenty percent slower, AI is acting as a research layer that extends consideration. A faster close at lower contract value points to AI pre-qualifying in ways that cap deal size. If the cohort shows up fine but loses at proposal more often than everyone else, a specific claim AI is making about you is surviving into late-stage conversations and costing you deals at the line.

What this signal can't tell you on its own is which AI claim is doing the work. That's what the supporting signals are for.

Supporting signals

The cohort test answers "is AI helping or hurting." The supporting signals answer "on which dimensions, propagated through which channels, surviving to which stage of the deal." Most teams should run two or three of these well rather than all of them badly. Start at the top.

1. Buyer-belief extraction from transcripts

Your sales calls are a continuous attribution dataset. When a prospect says "I read that you don't integrate with Salesforce," or "I heard your pricing is usage-based," or "someone told me you're SMB-only," they're reporting a belief about you that came from outside your controlled channels. Capture those statements as they happen, in whatever system your sales team already uses.

When the same belief starts appearing across unrelated prospects who didn't come from the same source, you're watching a narrative spread through the market. Increasingly, AI is the proximate origin. Cross-referencing the recurring beliefs against what AI currently says about you tells you whether the belief is AI-propagated and needs direct countering, or whether it's spreading through other channels and wants a different response.

The right substrate for this work is your existing call-recording stack. Tag the calls in real time or run a regular pass through the transcripts. Two layers of output:

• Beliefs AI is propagating that align with your positioning. AI is reinforcing the framing that's already working for you. Track these so you can double down on the content driving them.

• Beliefs AI is propagating that contradict your positioning. AI is actively costing you deals. Each one is a content priority: sales is saying X, AI is saying not-X, the action is the proof page that closes the gap.

2. Close-reason cross-reference

A particularly strong causal signal. Take what AI says about you (the strengths, the weaknesses, the dimensions it treats as important for your personas) and line it up against the reasons your buyers cite for winning or losing deals.

When buyers cite strengths on dimensions AI also describes favorably, AI is reinforcing what's working for you. When buyers cite weaknesses on dimensions AI warns them about, AI is actively costing you deals. And when buyers cite dimensions AI hasn't picked up at all, you have a narrative opportunity: the dimension matters to the buyer, and your side of the story hasn't propagated to AI yet.

Running this well requires two things that most teams already have half of. The first is structured close reasons in your CRM, tagged to specific attributes rather than a single free-text field. The second is a regular read on what AI currently believes about you across your personas. Unusual is what we built for that side; the move is available to any team willing to instrument both halves.

3. Lost-deal interviews

Retrospective conversations with lost prospects produce the cleanest qualitative attribution available. During an active deal, buyers have reasons to hide where their views came from. After the deal is closed-lost, those reasons disappear. Buyers describe the specific perceptions that drove them to a competitor, and they often name AI as the source.

One standard question to add to every lost-deal interview: "At any point during your evaluation, did anyone on your team consult an AI model to compare options? If yes, do you remember what it said?"

The data is qualitative and slow. Five to fifteen interviews per quarter is enough for most mid-market B2B companies. The output is a named list of perceptions costing you deals. Feed the list directly into the close-reason cross-reference, and use it to prioritize the positioning and proof work that comes out of the quarterly review.

4. Bot visits correlated with active deals

When `ChatGPT-User` or `Claude-User` visits your pricing page from a known prospect's IP range during their open sales cycle, you have direct evidence that AI was consulted by someone in that account mid-evaluation. Filter your existing bot-traffic data for IP ranges corresponding to active prospect companies, and overlay the timestamps against your CRM activity log.

This gives you per-deal evidence with timing alignment, useful for narrative attribution on specific deals. The CRO can show the board "here's the trace of AI activity on the deal we lost in week four to a competitor." That's the kind of artifact a CFO who is otherwise skeptical of AI attribution will actually look at.

5. Post-close call addition

A short AI-research question added to your standard post-close call (won and lost) gives you broad coverage at lower fidelity than the dedicated lost-deal interview. Higher volume, lower depth. Useful as the wide-net layer that the lost-deal interview deepens on selected deals.

A sample script: "We're learning a lot about how AI is shaping the way buyers research products like ours. Do you remember whether anyone on your team consulted ChatGPT or a similar tool during your evaluation? If yes, do you remember the gist of what it told you about us?"

6. Gong custom signal for AI mentions

For teams with a sales-intelligence platform configured to flag AI mentions automatically in call recordings, the signal trades depth for coverage. Every call gets a pass without an agent reading raw transcripts, and the frequency of AI mentions across the call corpus becomes a trackable metric. Lower fidelity than structured belief extraction, higher volume. The right substitute when transcript access is constrained.

Signals to ignore at this layer

Two metrics show up in conversion-attribution dashboards that don't earn their place.

UTM conversion rates in isolation. `utm_source=chatgpt` and the equivalents tell you what fraction of your direct AI traffic converts. Treated as a standalone read on AI's pipeline effect, the number is misleading. The click-through cohort is the lowest-conviction slice of AI-influenced buyers; higher-conviction buyers go straight to your domain or branded-search you, invisible to UTM. The UTM conversion rate is biased toward buyers already convinced enough to click. Useful as a floor estimate of the AI-direct cohort, not as a primary read on whether AI helps conversion.

Self-reported AI influence without a corresponding perception read. A lead-form question asking "did AI play a role in your evaluation" captures intent, not content. If you don't know what AI is actually saying about you, the self-report tells you a cohort exists but not what they were told. Pair the cohort flag with a perception read, or treat the cohort flag as a setup signal that the rest of the attribution work runs on top of.

What this looks like assembled

The work is quarterly. The shape of the review:

1. Pull the AI-affected cohort funnel comparison. Where does the cohort under-perform or out-perform baseline?

2. Pull the buyer-belief extraction from the quarter's sales transcripts. Which beliefs are recurring, and which match the AI framings the platform read has tagged?

3. Cross-reference structured close reasons against the AI framing read. Where do AI's strengths line up with wins? Where do AI's weaknesses line up with losses?

4. Pull the lost-deal interview themes. Whose verdict cites AI explicitly, and on what dimension?

5. Synthesize. For each persona, is AI an advocate or a detractor at the conversion stage, on which dimensions, and what specific content, positioning, or proof work moves it?

The output is a list of two or three perceptions worth moving in the next quarter, ranked by frequency in won and lost deals. Ship the content investments to move them, and re-measure next cycle.

Where this fits in the broader framework

This guide answers one of the four questions we use to frame the full attribution picture at Unusual. The Complete Guide to AI Attribution walks through the other three: what AI's perception of you actually is (the perception layer this conversion analysis sits on top of), whether AI is bringing you leads directly and indirectly, and whether AI itself is becoming a customer.

The conversion question is usually the one revenue teams care about most, and it's the question with the cleanest answer once the AI-affected cohort is tagged. Get the tagging in place, run the cohort comparison next quarter, and the rest of this guide becomes empirical work rather than narrative.