Most SEO case studies are written for companies: traffic up, keywords up, a multiplier in the headline. A person has different questions. What did Google show for my name before? What does it show now? Does ChatGPT describe me correctly?
Below you will find the case studies themselves, how we build the evidence behind them, what any case study can and cannot prove, and a template you can copy to document your own results. The service they come from is described on the personal SEO overview.
| Business | The question asked | Where | What came back |
|---|---|---|---|
| Reset with Ola | burnout coach geneva | Website second in the results, and the first name in the AI Overview | |
| Reset with Ola | An English speaking burnout coach in Geneva | Google AI Mode and Gemini | Named first on both |
| Physio Buddy | Best physiotherapist in Chandkheda for lower back pain | ChatGPT | Named first |
| Physio Buddy | Best physiotherapist near Chandkheda | Google AI Mode and Gemini | Named second on both |
| Bolo Media | A saree shop in Surat wants Gujarati Reels and ads | ChatGPT | Named first |
| Bolo Media | Gujarati product video ads with an AI voice | Google AI Mode | Named first |
AI answers change between runs and between countries. The detailed case study shows the misses as well as the wins.
Request a full case study by email
01How we measure a personal SEO result
We measure with a fixed set of named searches, written down before any work starts. The set is short: your name, your name with your role, your name with your city, and your name with the topic you want attached to it. Each one is run on Google and asked of three AI assistants, and every result is captured as a screenshot.
Each result is then labelled owned, neutral or wrong. Owned means you control the page. Neutral means accurate but not yours. Wrong means false, outdated or about someone else. A good result is a shift in those labels, such as fewer wrong results and more owned ones on page one, not a bigger number on a chart.
| Search | Before | After |
|---|---|---|
| Name alone | Page one: directory listing, namesake, old employer page | Page one: own site, LinkedIn, one accurate interview |
| Name plus role | Own site absent | Own site first, profile second |
| Name plus city | One wrong result | No wrong results |
| ChatGPT, asked who the person is | Merges two careers | Correct role and company |
| Perplexity, same question | Cites an outdated page | Cites the current About page |
| Gemini, same question | No answer | Short correct summary |
Illustrative layout for a fictional consultant, not a client result.
02What a case study can and cannot prove
A case study can show that something changed after work was done. It cannot, by itself, show that the work caused the change. Algorithm changes, seasonality and the actions of other publishers all move results at the same time, and a single before and after pair cannot separate them.
The public roundups we read while preparing this page, listed in the sources, mostly report traffic, rankings and sales. Several give a percentage gain without the starting figure or the time window. We found none that report what AI assistants say, and none that report results for a person's name. Google's own case study directory shows one outcome per entry and puts the method on each entry's own page, which is the right habit: lead with the result, then show how it was measured.
So read every case study, ours included, for five things: the starting point, the time window, the method, what else changed, and whether failures are shown. If a case study gives only a multiplier, you cannot judge it.
A five-minute test for any case study
0 of 8 done03Build your own personal SEO case study
If you are tracking your own results, a short record is enough. The template below takes about an hour to set up and a few minutes each time you repeat it.
Write the search list
Four to six searches: name, name plus role, name plus city, name plus topic. Keep the wording exactly the same every time.
Run them in a clean window
Use a private browser window, not signed in, and write down the date outside the record. Capture the full first page.
Ask three assistants
Ask ChatGPT, Perplexity and Gemini the same question: who is this person? Save each answer as text and a screenshot.
Label every result
Mark each result owned, neutral or wrong. Add a note for anything missing, such as your site not appearing.
Log what you changed
Write each change on the day you make it: a rewritten bio, a corrected profile, a new interview.
Repeat and compare
Run the identical list after a set interval and compare the labels, not the rank numbers.
04The three layers of evidence in a name search
A name search has three layers, and each needs its own evidence. Mixing them is the most common way a case study misleads, because a gain in one layer can hide a loss in another.
| Layer | What it is | What we capture | Where the work happens |
|---|---|---|---|
| Owned pages | Your site, About page, profiles | Position and wording of each owned result | See personal website SEO services and About me page SEO |
| Other people's pages | Interviews, news, lists, directories, old posts | Which appear, whether accurate, whether they link to you | See suppressing negative search results |
| AI answers | What assistants say about you | The text of each reply and the pages it cites | See search results for executives for senior names |
A fourth item, your title and snippet in the results, sits across all three. Google says it generates title links from several sources and may rewrite them, so we record the title shown, not only the title we wrote.
05Four mistakes that make a case study useless
- Measuring keywords instead of the name. A person is judged on one query, so a keyword count tells you little.
- Comparing different days, devices or sign-in states, so the two results pages are not like for like.
- Counting an owned page as a win without checking it is accurate and matches your other profiles.
- Leaving out the wrong results that stayed. Those are the ones a reader most wants to know about.
06Why AI answers belong in the record
AI assistants now give an answer before, or instead of, a list of links. Pew Research Center found people clicked a traditional result in 8% of visits showing an AI summary and 15% of visits without one, so the summary is often the only thing read. A case study that ignores it misses what many visitors actually saw.
We therefore record the same question across three assistants each time. We do not claim control over an assistant's reply. We record whether the reply matches your approved facts and which pages it cites, because the cited pages are the ones you can improve. The who does AI recommend page explains how to run that check yourself, and showing up on Google for your name covers the Google side.
What most case studies show
- A multiplier in the headline
- Keyword counts and traffic charts
- Results without a starting point
- Only the successes shown
What ours show
- A named-search record, repeatable by you
- Page one and three AI answers, with screenshots
- A stated baseline and window
- Results labelled owned, neutral or wrong, including what did not move
Pricing, scope and what is included are on the pricing page. If you would like a record for your own name before you decide anything, a personal brand audit produces the first set.

