AEO Pipeline Attribution

Most of what's sold as "AEO" in 2026 is monitoring. You pay $399 a month to see where your brand appears in ChatGPT answers for a set of tracked prompts. You get a share-of-voice chart. You don't get a dollar number.

Pipeline attribution is the opposite side of that line. The CMO who has it can walk into a board meeting and say "AEO generated $412,000 of pipeline this quarter, here are the three deals, here's the CAC payback." The CMO who doesn't has to hedge.

This page is the definition, the math, the six-step citation-to-closed-won chain, and the reason AEO pipeline attribution is a separate category from AEO monitoring. It references our CFO's Guide to AEO Budget Defense and the Lantern vs Profound comparison for readers who want to go deeper.

Why AEO needs its own attribution model

Three facts make standard web attribution useless for AI-referred demand:

1. 70.6% of AI traffic lands in GA4 as Direct. ChatGPT, Perplexity, and Claude sessions typically open your site via copy-paste or a link-tap from a mobile app — neither produces a reliable referrer header. Loamly measured this in Q1 2026; your own GA4 will show something similar if you segment carefully.
2. Last-click attribution over-credits the channel that closed, not the one that discovered. In the traditional funnel, the ChatGPT citation that first introduced your brand ends up credited to whatever the buyer clicked last — usually branded paid search, direct, or a gated demo form.
3. The demand-creation moment has moved inside private LLM chat windows. The conversation where a buyer decides whether your category is real, whether your product solves their problem, and whether you or a competitor is worth evaluating now happens inside a ChatGPT window no analytics tool can read. There are no footprints to follow — unless you build them yourself.

This is the exact pattern one B2B SaaS founder described: "Five new leads came in yesterday. HubSpot attributed them to Direct, Paid Search, and Organic Search. But when you asked the leads directly, all five said: I found you on ChatGPT."

Pipeline attribution for AEO exists to close that gap.

AEO monitoring versus AEO pipeline attribution

The distinction is categorical. Monitoring tells you where you are cited. Attribution tells you whether those citations produced pipeline. The table:

This is the line the audience brief describes in one phrase: monitoring, not attribution. It is the single-sentence takedown of every AEO dashboard that can't close the loop to revenue.

The six-step citation-to-closed-won chain

AEO pipeline attribution works by instrumenting each link of the chain from an answer-engine citation to a closed-won opportunity. The six steps:

Step 1. Citation detection

An AI engine (ChatGPT, Perplexity, Claude, Gemini, AI Overviews) answers a tracked prompt and cites your brand — either a direct link, a quoted passage, or a named mention. Detection is done by sampling each prompt N times per engine per day (typical range: 5–50 samples per prompt per engine per day). Citations are logged with prompt, engine, rank, surrounding context, and timestamp.

Step 2. Session identification

A user reads the answer, clicks through (or copies a URL and pastes it into a browser), and lands on your site. Because the referrer is unreliable, identification relies on a combination of signals: UTM parameters on outbound links inside citations you influence, session timing against known citation-serve windows, session entry pages commonly cited by engines, and — most importantly — a self-reported attribution form field on every inbound form ("How did you first hear about us?").

Step 3. Contact resolution

The session becomes a contact record in HubSpot (or Salesforce) through a form submission, gated download, or demo request. On creation, the contact record is written with one or more AEO source properties: aeo_first_touch_engine, aeo_first_touch_prompt, aeo_first_touch_timestamp, aeo_self_reported_source. These properties are the CRM-side instrumentation that makes everything downstream possible.

Step 4. Opportunity association

The contact becomes an MQL, then an SQL, then an opportunity. In HubSpot, AEO source properties propagate from contact to company to deal via standard property sync rules. Every opportunity now carries a flag that answers one question: was this deal influenced by an AEO touchpoint at any point in the buyer journey?

Step 5. Stage-weighted roll-up

Each AEO-influenced opportunity is rolled up into pipeline using a stage-weighted probability model: an opportunity in Discovery weighs in at 10%, Proposal at 40%, Commit at 80%, and so on — using the customer's own HubSpot deal stage configuration. AEO-influenced pipeline dollars are reported monthly against the total pipeline created in the same period.

Step 6. Closed-won credit and ROI report

When an AEO-influenced opportunity closes won, it is credited back to the first AEO touchpoint (first-touch) and — if multi-touch is enabled — distributed across all intermediate AEO citations on a decay model. The closed-won ARR becomes the numerator for AEO ROI: program cost (tool + content + attributable headcount) divided into AEO-influenced closed-won revenue gives the AEO CAC payback in months. This is the number that ends up in the quarterly memo to the CFO.

The math, written out

Two formulas do the work. Both must be shown in any CFO-defensible memo.

Both formulas expose their assumptions. That is the point. A CFO trusts a transparent model with stated assumptions far more than a tool-generated number whose sourcing can't be inspected. See the CFO's Guide to AEO Budget Defense for the memo template, the objection cheat sheet, and the scorecard.

First-touch versus multi-touch in AEO

A buyer's journey through AI answers rarely involves a single citation. They ask ChatGPT about the category, get four tools named, check Perplexity to corroborate, read a comparison page, go quiet for two weeks, come back to AI Overviews with a more specific intent query, and then land on a pricing page. How you attribute credit across those touches is the first-touch-vs-multi-touch decision.

The working guidance for B2B SaaS in 2026:

• First-touch for renewal defenses and category-formation claims ("AEO introduced this buyer to our category"). Simple, defensible, and the CFO understands it.
• Multi-touch with a time-decay weighting for internal prompt and content prioritization ("which of our 40 tracked prompts produced the most pipeline"). Needed because multiple citations across engines typically influence the same deal.
• Both, in parallel is the mature setup. Lantern reports first-touch in the monthly CFO PDF and multi-touch in the weekly operational dashboard.

For a deeper dive into the math: see our AEO attribution methodology page.

Why this is a new category

AEO pipeline attribution is not a feature bolted onto a monitoring tool. It is a separate product category because it requires three capabilities that monitoring tools, by design, don't have:

1. Native CRM integration with write access to contact, company, and deal objects. Every AEO monitoring tool on the market ships a standalone dashboard. None of them write source properties to HubSpot or Salesforce in V1.
2. A stage-weighted pipeline model that honors the customer's own deal stage configuration — not a generic "funnel." CFOs don't accept generic funnels; they accept their own numbers, run through their own probabilities.
3. An attribution team's mental model, not a SEO team's. The underlying questions AEO pipeline attribution answers (lookback windows, first-touch-vs-multi-touch, credit distribution, fractional assignment) are attribution questions, not citation-tracking questions. The companies that built the first-generation AEO monitoring tools came from the SEO world. That architecture is not easily retrofitted.

The category is also new because the buyer is different. Monitoring tools are sold to SEO leads and content leads — the operational owners of AEO. Pipeline attribution is sold to CMOs and Heads of Growth — the budget owners. Different buyer, different KPIs, different purchase cycle, different renewal math.

What a CFO-defensible AEO pipeline number looks like

The test is simple: can you hand a single number to your CFO, have them ask two or three questions, and be able to answer every one from the underlying model — without a vendor call, without a "let me get back to you," without hedging?

A CFO-defensible number has four properties:

• It is in dollars, not citations. "$412K of AEO-influenced pipeline in Q1" > "share of voice up 12%."
• It exposes its assumptions. Stage-weighted probability disclosed, lookback window disclosed, source-property definition visible.
• It lives in the CRM, not a vendor dashboard. The CFO can ask Marketing Ops to pull a list of the underlying opportunities in HubSpot and see the same number the CMO is reporting.
• It benchmarks against the rest of the marketing mix on the same denominator. AEO CAC payback vs paid CAC payback vs SEO CAC payback, all in months, all on the same ACV assumption.

How Lantern operationalizes this

Lantern is AEO pipeline attribution for B2B SaaS. We ship the full six-step chain: citation detection across ChatGPT, Perplexity, Claude, Gemini, and AI Overviews; session and contact resolution via form-level self-reporting plus UTM pattern matching; source properties written natively to HubSpot contact, company, and deal objects; stage-weighted roll-up using the customer's own HubSpot deal stage probability model; first-touch credit in the monthly CFO PDF and multi-touch in the operational dashboard; and a CAC-payback calculation on closed-won ARR.

HubSpot is the V1 integration (live). Salesforce is V1.5. Pricing is $99/mo or Enterprise — two tiers, no hidden prompts-per-seat math. The monthly deliverable is a PDF ROI report a CMO can forward to the CFO without edits.

For a direct comparison to citation-monitoring tools, see Lantern vs Profound. For the renewal-defense memo, see the CFO's Guide. For the HubSpot-specific implementation: AI search attribution in HubSpot.

Common mistakes

• Tracking citations instead of pipeline. Citation count is operationally useful, not CFO-useful. Demote to an appendix metric.
• Running AEO attribution in a separate tool from your CRM. The number you report has to live inside HubSpot (or Salesforce), where the CFO can audit it. A standalone AEO dashboard will lose every renewal fight.
• Using last-click for AEO. AEO is a discovery channel. Last-click attribution will under-credit it structurally, then the channel will look unprofitable, then the budget will get cut. Use first-touch (with lookback) or multi-touch.
• Tracking every possible prompt. Most B2B SaaS categories have 7–12 prompts that drive 80% of AEO pipeline. Broader prompt coverage adds noise, not signal. Audit monthly; cut aggressively.
• Not writing self-reported source on every form. The cheapest, most CFO-defensible attribution input is a free-text field on every inbound form asking "How did you first hear about us?" Contacts who write "ChatGPT" or "Perplexity" are your ground truth. Skipping this step is the single largest attribution leak in most B2B SaaS stacks.

The question this page answers

When your CEO asks "how are we showing up in ChatGPT — and did that generate any pipeline?", the CMO with AEO pipeline attribution has a two-sentence answer. AEO-influenced pipeline for the quarter in dollars. AEO CAC payback in months, benchmarked against paid and SEO. Everything else — share of voice, citation count, prompt coverage — is supporting detail.

That's the category. That's why pipeline attribution is distinct from monitoring. That's what Lantern ships.