Get a baseline, then work the issues list. On Saturday evening, October 24, 2026, sociallysmashing.com scored 72.2 on an AI readiness audit. By Sunday evening, the score was 92.5. The method: run the audit, fix one category at a time, and recheck. A score is a signal, not a guarantee.
This is a record of what I did, in what order, and what the numbers did. If you want to run the same process on your own site, the method is here.
sociallysmashing.com. AI readiness score, Saturday evening, October 24, 2026, to Sunday evening, October 25. A gain of 20.3 points, with a site health score of 92.8.
nicolejolie.com. The same process, the same weekend. The gain was smaller in points.
nicolejolie.com open issues. 543 when I started, 208 by my last check that Sunday. 335 fewer open issues.
A client's contracting site. Open issues, Sunday evening, 7 PM to 9 PM. Part of that work was placing 60 of the business's existing Google reviews across 20 pages of its site.
An AI readiness score does not tell you where you appear in a list of results. It tells you how prepared your site is to be read, understood, and cited by AI engines like ChatGPT, Perplexity, Claude, and Google's AI Overviews.
| Category | What it checks |
|---|---|
| Crawler access | Can AI bots reach and read your pages? |
| Content quality | Do your pages answer questions clearly and completely? |
| Structured data | Have you labeled your business, services, and content in machine-readable formats? |
| Brand authority | Do external sources (directories, reviews, press) confirm what you say about yourself? |
| Technical hygiene | Are pages indexed, free of errors, and internally linked? |
The "open issues" number is separate from the score. It is a count of specific problems the audit found across your site. A high score with many open issues means the overall structure is solid but individual pages or elements still need attention. A low score with many open issues means the problems are more fundamental.
The score is the summary. The issues list is the work order.
The first task was establishing where each site stood before I changed anything. This step is easy to skip, and skipping it is why most people can't tell later whether their changes made a difference.
I ran AI readiness audits on sociallysmashing.com and nicolejolie.com back to back. The audits flagged issues across multiple categories for both sites.
Your check: score and date written downOn Sunday I worked through the Critical and High issues site-wide, including structured data. These take longer than content edits because they mean going page by page and checking specific technical conditions. They don't require a developer if you're comfortable with your site's backend.
Your check: name your most severe issueThe audit had flagged 200 open issues. In two hours, that number dropped to 111. One part of the work was placing 60 of the business's existing Google reviews across 20 pages of its site.
This addresses the brand authority category. AI engines look for external confirmation of what a business claims about itself. Reviews on individual service or location pages give AI engines specific, verifiable signals about what the business does and what customers say about it.
Your check: reviews on your service pagesTwo hours is a short window. It was possible because the audit had already done the diagnostic work. I wasn't guessing what to fix. I was working through a prioritized list.
The pattern that works: Run the audit. Fix one category at a time, starting with the highest severity. Recheck. The audit tells you where to look; the work is in the doing.
The most consistent finding across both sites was that key pages weren't opening with a direct answer. They had taglines, mission statements, and introductions that described the business without answering a customer question.
Google's guidance for generative AI features is specific: AI systems extract passages from pages, not whole pages. A passage needs to make sense on its own to be useful to an AI engine building an answer.
So I rewrote the opening paragraph of several pages so the first 50 to 70 words answer the most likely customer question. Not a tagline. Not a welcome message. What does this site (or business) do, for whom, and why does it matter?
Vague headings like "Our Services" give AI engines nothing to work with. Descriptive headings like "What does an AI visibility audit include?" give a clear signal about what the section answers.
According to HubSpot's research on answer engine visibility, structuring content around specific questions is one of the most reliable ways to improve citation rates over time.
Added or corrected on the most important pages.
Fonts on nicolejolie.com set to load without blocking the page.
The weekend gives you a baseline and a first round of fixes. What comes next depends on which category scored lowest on your audit.
Your pages probably aren't answering customer questions clearly enough. Write one FAQ page for your most important service, with question-based headings and a direct answer under each one. Keep each answer to 50 to 100 words, and make sure it makes sense read on its own.
AI engines look for external confirmation of what you say about yourself: consistent business listings, reviews on relevant platforms, and mentions from third-party sources. Make sure every field on your Google Business Profile is complete, accurate, and matches what your website says about your business name, address, services, and hours.
Schema markup is specific and easy to get wrong in ways that don't show up visually but do affect how AI engines read your pages. A developer or an AI visibility specialist can implement it correctly in less time than it would take to learn it from scratch.
Look for accidental noindex tags, content that only loads with JavaScript, and slow load times or server errors. Your audit will flag most of these. Fix the ones you can without a developer first, then decide whether the rest need outside help.
An AI readiness audit score is a signal, not a guarantee. A higher score means your site is better structured for AI engines to read, extract, and reference. It doesn't promise that ChatGPT will name your business in every answer. What it tells you is whether you're making it easy or difficult for AI systems to understand what your business does and who it serves.
One weekend is a starting point. You now have a specific baseline, a list of what you changed, and a comparison point. Run the audit again in 30 days. Note what moved and what didn't. Each cycle, you build a clearer picture of what's affecting your AI visibility and what's worth fixing next.
Have Nicole Jolie do this for you. She runs the prompts, reads the reports, and writes the pages in your buyers' questions. You answer questions and approve pages.
See the VIP programsociallysmashing.com/recommended/vipShort, practical, and built to be used the same day.
One AEO fix each week, plus first access when the community opens.
Join the newsletterAn AI readiness score is not a search ranking. It tells you how prepared your site is to be read, understood, and cited by AI engines like ChatGPT, Perplexity, Claude, and Google's AI Overviews. Most audits break that into crawler access, content quality, structured data, brand authority, and technical hygiene.
The open issues number is separate from the score. It is a count of specific problems the audit found across your site. The score is the summary. The issues list is the work order.
No. An AI readiness audit score is a signal, not a guarantee. A higher score means your site is better structured for AI engines to read, extract, and reference. It does not promise that ChatGPT will name your business in every answer.
sociallysmashing.com went from 72.2 to 92.5, with a site health score of 92.8. nicolejolie.com went from 82.9 to 93.7, and its open issues went from 543 to 208 by the last check that Sunday. A client's contracting site went from 200 open issues to 111 in two hours on Sunday evening.
Run the audit again in 30 days. Note what moved and what didn't. Your next project depends on which category scored lowest on your audit: content quality, brand authority, structured data, or crawlability.