GEO pricing in 2026
What audits, programs and retainers cost, what you should get at each price, how to compare quotes, and the prices I charge, on the page.
By Daria Dovzhikova · September 2026
The short version
- Three tiers: automated audits at $99 to $499, consultant audits at $1,500 to $5,000, retainers at $2,000 to $8,000 a month for mid-market.
- The price is set by prompt count, engines covered, whether a person reads the answers, and whether fixes are included.
- Every band on this page is an asking price published by a seller. Nobody publishes what buyers paid.
- Mine: one Growth Pilot at $5,000 for four weeks, baseline included; Continuation at $2,500 a month after.
Where the numbers come from
Every published GEO price list I have found is written by a seller. The Digital Elevator guide (updated July 2026) puts freelance projects at $150 to $2,000, productized packages at $500 to $2,500 a month, mid-market retainers at $2,000 to $8,000 a month and enterprise programs at $10,000 to $30,000 a month and up. WebFX (May 2026) gives $1,500 to $5,000 a month for small businesses, $5,000 to $25,000 for mid-sized, $5,000 to $50,000 per project, and starts its own service at $3,000 a month. Veza Digital sells a single audit at $4,500 fixed: 50 to 100 prompts across six platforms, 30 days of tracking, a 30-page report and a 90-day plan.
Read all of that as what sellers ask. I have not found a survey of what buyers paid, and one guide cites the other. Treat the bands as a map of the market, not a price you are owed.
The four price tiers
| Tier | Price | What you get | What is missing |
|---|---|---|---|
| Automated audit | $99 to $499, one-time | A script runs 5 to 30 prompts through one to five engines and returns a PDF. Nobody reads the answers. Useful for a first look at whether you appear at all. | Competitor scoring, the pages the engines cite, a fix list anyone can act on. |
| Consultant audit | $1,500 to $5,000, one-time | A person writes the prompts, reads every answer, scores you against the competitors the engines name, lists the cited pages, and ranks the fixes. Two to six weeks. | The fixes themselves. An audit ends at the list. |
| Program, fixed scope | $5,000 to $50,000, one-time | The audit plus the fixes shipped: extraction gaps, canonical positioning in llms.txt and JSON-LD, comparison pages, and a re-measurement at the end. | Ongoing tracking after the closing read. |
| Retainer | $1,000 to $8,000 a month, mid-market | Monthly re-measurement across the engines and a steady stream of fixes as the answers move. Enterprise programs run $10,000 to $30,000 a month and beyond. | A clear end. Ask what the monthly number is measured on. |
The gap between the first two tiers is not margin. An automated audit tests whether your domain string appears in an answer. A consultant audit records who was named instead of you, which page the engine cited, and what to change first. One is a smoke test. The other is a diagnosis.
What drives the price
| Driver | Why it moves the price |
|---|---|
| Prompt count | Ten prompts tell you where you stand on ten questions. Fifty tell you where you stand in the category. Prompt count is the single biggest cost driver, because each prompt is read by a person at the consultant tier. |
| Engines covered | ChatGPT, Claude, Gemini and Perplexity answer the same question differently and cite different pages. Google AI Overviews and AI Mode are a further two. Every engine added multiplies the reading. |
| Human reading | An automated check tests whether your domain string appears in the output. A person records who was named instead of you, which page the engine pulled from, and whether the recommendation was conditional. That difference is most of the gap between the automated tier and the consultant tier. |
| Fixes shipped or report only | A report costs one price. A report plus the fixes, plus a re-measurement that proves they worked, costs three to six times that. The second thing is what moves the number. |
| Re-measurement | Anyone can sell a baseline. The closing read, on the same prompts and the same engines, is what turns an audit into evidence. Ask whether it is included. |
What I charge
Growth Pilot, $5,000, four weeks. Week one is the baseline: 48 unbranded prompts across ChatGPT, Claude, Gemini and Perplexity, every answer read by a person and scored against the competitors the engines name instead of you, with the pages they cite. The fixes it names become two experiment cards, decided every Friday and re-measured on day 28 on the same prompts. Method on the AI search visibility page.
Continuation, $2,500 a month. Only after a pilot. The same read, every month, with one new card a month. Keeps the number moving as the answers move. Every price and every term is on the pricing page.
Against the market bands: the pilot sits at the low end of the project band with the audit included, and Continuation sits under the mid-market retainer median. The scope is written down so you can compare it with any other quote line by line.
How to compare quotes
Six questions. If a seller cannot answer one of them in a sentence, the price on the quote is not comparable to anyone else's.
- How many prompts, and who wrote them? Buyer questions, or brand searches padded to look like a category?
- Which engines, by name? A quote that says AI search without naming the engines is a quote for one engine.
- Does a person read the answers? Ask to see a sample scored answer from a previous audit.
- What is in the fix list: pages to change, or advice?
- Is there a closing re-measurement on the same prompts? If not, you are buying a snapshot, not a result.
- If you continue past the audit, is it credited, or is the audit the whole bill?
Before you pay anyone, run the free visibility check. It is the automated tier, done honestly: it tells you whether you appear at all, and nothing more. If the answer is no on every prompt, that is the whole audit, and it was free.
Questions
How much does a GEO audit cost?
Automated audits run $99 to $499 and return a PDF nobody read. Consultant audits run $1,500 to $5,000 and include a person reading every answer, competitor scoring, cited pages and a ranked fix list. One named seller charges $4,500 for 50 to 100 prompts across six platforms over six weeks. Mine is a 48-prompt, human-read baseline, and it runs in week one of the Growth Pilot ($5,000, four weeks) rather than as a separate audit.
What does GEO cost per month?
Published retainer bands for mid-market companies run $2,000 to $8,000 a month, with monitoring-only tiers from $1,000 and enterprise programs from $10,000. After a pilot, Continuation at $2,500 a month re-measures the same prompts on the same four engines every month.
What is the difference between an audit and a program?
An audit ends at a list of fixes. A program ships the fixes and re-measures at the end. Here that is the Growth Pilot, $5,000 over four weeks: the baseline in week one, the fixes as two experiment cards, and a day-28 read on the same prompts.
Why do GEO prices vary so much?
Four things: how many prompts are tested, how many engines, whether a person reads the answers or a script checks for your domain string, and whether the price covers fixes or only a report. An automated audit and a consultant audit are not the same product with different margins. They are different products.
Is GEO pricing separate from SEO pricing?
Mostly not, and it should not be. Most of the fixes that move AI citations are pages, structure and clarity, which is the same work an SEO program does. What is separate is the measurement: which engines name you, on which prompts, against whom. Pay for that measurement once, then fold the fixes into whatever program runs the site.
Are these prices what people actually pay?
The market bands on this page are asking prices published by sellers, not closed deals. I have not found a survey of what buyers paid. My own prices are what I charge, and they are on the pricing page.
Related guides
The Growth Pilot
Four weeks, two real growth experiments, a decision every Friday.
The 48-prompt baseline above is week one of the pilot; the fixes it names are the two cards. For developer-tool and AI-product founders who don't have a growth lead yet. $5,000 fixed, half at kickoff. How a pilot runs · What an experiment card looks like
Book a 30-minute callWant the number before you decide?
Thirty minutes. I will tell you which tier you need, and whether you need one at all.