How to Report AEO Progress in a Quarter With Zero Attributable Deals

The worst version of this meeting is the one where you open with an apology. You ran answer engine optimization for a full quarter, the citations are climbing, and your CRM shows exactly zero deals sourced to it. The instinct is to either overclaim, tagging ambiguous Direct traffic as AI-sourced to produce a number, or to go quiet and hope nobody asks. Both lose the room. The CFO who catches one invented attribution stops trusting every figure after it, and the marketer who reports nothing looks like they spent a quarter on nothing.

There is a third path, and it is the only durable one: report the signals that genuinely moved, explain in plain terms why revenue has not caught up yet, and show the lag as a feature of the measurement problem rather than a failure of the work. This is the quarter-with-zero-deals report done honestly, section by section, so a skeptical finance partner leaves believing the line is real even though the revenue column still reads zero.

Why zero attributed deals is the expected result, not the alarm

Start the report by naming the mechanism, because the silence in your CRM is a measurement artifact before it is anything else. AI assistants almost never pass a clean referrer. Clicks from the ChatGPT app, from logged-in sessions, and from privacy-restricted contexts arrive with no source header, so analytics defaults the session to Direct. The scale of this is not marginal. An analysis of AI referral sessions and multiple 2026 traffic studies put the share of AI visits that land in analytics as Direct in the range of 35 to 70 percent, with some ChatGPT-specific samples higher still. Your measured AI channel is a floor, not a total.

That means the absence of tagged deals is telling you almost nothing about whether AEO drove pipeline. A buyer can ask ChatGPT for tools in your category, see you named, click through, land as Direct, browse for two weeks, and convert through a branded search your model credits to organic. Every step of that journey is invisible to last-click. The full mechanics of why this revenue hides, and the three-signal method for surfacing it, are laid out in how to prove AEO pipeline when the buyer leaves no referrer. Open the report here so the zero reads as a known limitation of attribution, not a verdict on the work.

Lead with the four indicators that move before revenue

With the lag named, report the things that actually changed. Leading indicators exist precisely for quarters like this one, when the outcome you care about has not arrived but the signals that predict it have. Four of them belong at the top of the report.

Citation share is your share of voice inside AI answers: across a fixed set of category prompts, how often you are named versus competitors. It is the closest AEO equivalent to keyword rankings, and it is the number the whole program exists to move.

Citation velocity, the month-over-month change in that share, is the single most predictive figure you have. Citations rise first; traffic and pipeline follow. A flat citation share that suddenly ticks up in month three is the leading edge of everything downstream, and reporting velocity separately from the absolute number is what lets a CFO see momentum before it shows up in dollars.

Prompt coverage is the breadth of questions where you appear at all. Going from cited on 3 of 40 tracked prompts to cited on 14 of 40 is real progress even if your average position is still low, because coverage is what you convert into share next quarter.

Branded search lift is the bridge indicator. When AI answers name you in zero-click interactions, the downstream fingerprint is a rise in people searching your brand directly. A citation share that climbs while branded search stays flat is a warning that the mentions are not landing; the two rising together is the strongest available evidence that AEO is building demand you will bank later. The leading-indicator hierarchy, and which metrics predict pipeline versus which merely flatter, is worked through in which AEO metrics actually predict pipeline.

Put the lagging number next to the leading one, with the delay stated

The mistake that sinks these reports is showing the leading indicators alone, which reads as changing the subject away from revenue. Do the opposite. Put the lagging figure, influenced pipeline or sourced revenue, right beside the leading ones, and label the expected delay between them explicitly.

The line that works sounds like this: citation share rose from 4 to 11 percent this quarter, branded search followed with an 18 percent lift six weeks later, and influenced pipeline is where we expect the signal to surface next quarter based on our sales cycle length. You are not hiding the zero. You are placing it in a sequence and dating when it should change. A finance partner can accept a zero that comes with a mechanism and a date far more easily than a zero you skate past.

Here is the reporting frame that keeps the two honest and legible side by side:

The numbers in the table are illustrative; the structure is the point. Reporting leading and lagging figures in one view, with the lag named, is what the outcomes-first approach to AEO measurement argues for, and it is the difference between a report that manages expectations and one that quietly sets you up to be blamed when revenue does not appear on the CFO's preferred timeline.

Recover what attribution you can, without inventing any

Zero attributed deals does not mean zero attribution effort. It means the default tooling missed them, and there are legitimate ways to recover some signal before you concede the number is truly blank. None of these fabricate a source; they surface real ones the referrer stripped.

The fastest is a custom channel group that catches the AI traffic that does pass a referrer. Building a source regex that matches chatgpt, perplexity, claude, gemini, and copilot, then reporting on it as its own channel, pulls a real subset of AI visits out of the Direct bucket in about ten minutes. It is still an undercount, a floor rather than the true total, but a measured floor beats an assumed zero. The setup, and how to tie it through to pipeline, is in how to track AI referral traffic in GA4 and tie it to pipeline.

The higher-fidelity recovery is a self-reported attribution question at conversion: a "how did you first hear about us" field with an explicit AI-assistant option. Because the app-based and logged-in traffic that analytics cannot see is exactly the traffic a buyer can remember and name, self-reported data routinely surfaces AI-sourced deals that every automated model missed. Report both numbers, the channel-recovered floor and the self-reported count, and label each for what it is. What you must not do is take ambiguous Direct traffic and relabel it AI-sourced to manufacture a figure. One invented attribution, caught once, costs you the credibility of the entire report.

Frame the quarter against a benchmark so zero has context

A zero in isolation looks like failure; a zero measured against where brands actually start looks like the expected first step. Close the numeric section by placing your citation share on the real curve. Category leaders hold roughly 35 to 70 percent AI share of voice, while new entrants typically start at 2 to 10 percent and climb from there. If you moved from 4 to 11 percent this quarter, you are not stalled, you are on the entrant's ramp, and the benchmark makes that legible instead of leaving the CFO to assume 11 percent is disappointing. The full set of 2026 benchmarks by platform, and a method for setting a defensible target off them, is in what is a good AI citation share. A target on the report turns this quarter's number from a lonely data point into the first plotted position on a trajectory.

Show the machine, so the wait reads as process

The last section of a zero-deals report should answer the unspoken question: if revenue has not moved, what exactly are we paying for. Answer it by showing the operating loop, not by promising a number. The loop is legible: measure where engines name competitors and not you, produce content structured to be the answer to the exact prompts buyers ask, and track citation share as it climbs. That model is what how OnlyAEO works describes, and the always-on production that keeps feeding engines answer-shaped pages is the job of the AI Feed Engine. A CFO who can see the mechanism understands that the citation line moving is the product working, and the revenue line is a lagging readout of the same motion.

If you want a named arc rather than an abstraction, the FastTrackr AI case study shows a brand moving from unnamed to cited by rebuilding pages around real buyer questions, which is the same starting position as a brand posting its first zero-deals quarter. And for a concrete action that starts widening the surfaces engines can read this week, a free llms.txt generator produces a real file in minutes, so the report can point to work already in motion rather than work merely planned.

A quarter with zero attributed deals is not a quarter with nothing to report. It is a quarter where the leading indicators are the story and the lagging one is a dated promise. Report the citation share and its velocity, name the referrer problem in plain terms, recover the attribution you honestly can, and place the whole thing against a benchmark. Do that and the CFO reads the zero the way you do: as the last quiet moment before the line that already moved reaches the number they watch.