Questions agencies ask before their first audit

Traceponse is an AI search visibility audit for local and service businesses, built for marketing and SEO agencies. We ask ChatGPT, Perplexity, Gemini and Google AI Mode the questions your client's customers ask, record who gets recommended and cited, and turn that into a prioritized, white-labeled report with the evidence behind every fix.

AEO (answer engine optimization), also called GEO, is about how AI answer engines describe and cite a business. SEO is about where a page ranks in a list of links. An AI engine writes one answer and cites a few sources, so a business that isn't named or cited is invisible for that question. Traceponse is the measurement and diagnosis side: where the business stands, and what to fix.

A ranked list of fixes with the evidence behind each one. For every query a competitor won, you see who was cited and which specific facts their page states that your client's page doesn't. You also get the third-party sources AI engines lean on in your client's industry, a check of the Google Business Profile, map pack and name, address and phone details, a technical baseline (AI crawler access, llms.txt, structured data), and a developer brief with ready-to-paste schema and a directory checklist.

ChatGPT, Perplexity, Gemini and Google AI Mode. We don't currently check Claude.

Local and service businesses get a dedicated check on top of the AI answers: their Google Business Profile and map-pack listing, and whether their name, address and phone match across their site and listings. Queries name the city, never "near me", because engines can't resolve "near me" for a simulated searcher.

We focus on US businesses today. The address and phone checks are US-only.

A prompt tracker watches a list of prompts over time. Traceponse audits one client in depth.

The difference matters for reliability. AI engines give different answers to the same question from one run to the next, so a single response proves nothing. You have to ask each question multiple times before you can say how confident you are in a result. Many low-cost trackers don't tell you how many times a prompt was run or how confidence is calculated, and a visibility number that can't show that shouldn't go in front of a client. Ask any vendor before you rely on its numbers.

We run each question more than once, state our confidence in every finding, and name the exact fact a competitor's cited page has that your client's page lacks.

You can ask ChatGPT a question yourself, but one answer is a sample of one. A result you can show a client takes real questions for each buyer type, run across four engines and repeated enough times to separate a pattern from noise. It also takes every cited source recorded and compared with competitors' pages and the business's Google listing. That is possible by hand, but it takes hours per client, and without repeating each question you can't say how reliable the result is. We do that work and hand you a report ready to present.

No. Traceponse is a point-in-time audit. When you want to show what changed, for example after a fix ships, we rerun the same questions as a follow-up audit. If you need daily monitoring, a prompt tracker is the right tool. Be careful with its daily numbers, though: a daily figure based on one run per prompt mostly measures noise, not change.

No, and nobody honest can. We measure what AI engines say today and show what to fix. We don't promise a result or a timeframe, and we don't claim why an engine picked one source over another. We report what the winning page has that your client's page lacks.

They do change, so a single answer isn't evidence. We run each question more than once and report how confident we are in each finding. Results are a point-in-time sample, so if too few queries succeed, the report says so instead of presenting thin numbers as solid.

We don't share it. Each audit is kept internally as a baseline, so that a future follow-up audit can compare like for like.

Request one with a client's website and your work email. We email you to confirm the client site, and send the audit within 48 hours. Each agency gets one free audit.

Yes. It's written so you can present it to your client as your own work.

A client's website address and your work email. If there's a question your client wants to win, such as "best family dentist in Boise", add it and we'll build the audit around it.

$149 per audit, with a 5-pack and a monthly plan if you run audits regularly. There are no seat licenses and no dashboards to pay for. You set your own price to your client.

We rerun the same personas and questions against a fresh crawl, so you can show your client what was resolved, what persists and what's new. We report what changed between the two audits, not which fix caused it.

No. The audit diagnoses and prioritizes. It includes a developer brief your team or your client's developer can act on.