How to Fix Low AI Share of Voice vs. Competitors
AI share of voice measures how often AI assistants such as ChatGPT, Perplexity, Gemini and Google AI Overviews mention your brand in answers to buyer questions, compared with your strongest competitor. This guide shows you how to find the prompts where competitors get named and you don't, why that happens, and how to close each gap. You will finish by re-running the same prompts to check that the gap has narrowed.

Why it matters
A growing share of product research starts and ends inside an AI answer. If an assistant lists three tools for "best invoicing software for freelancers" and you aren't one of them, you never reach the shortlist, and no classic ranking report will show that loss. The audit compares your average mention rate across AI engines with the mention rate of your strongest competitor.
You pass if you are within 10 percentage points of them, get a warning between 10 and 30 points behind, and fail beyond that. Closing the gap builds on itself, because assistants tend to reuse the sources and brands they already trust.
What you need
Step-by-step fix
Build a prompt set that mirrors buyer questions
Write 30–50 prompts that a buyer would type before choosing a product like yours. Leave your brand name out of all of them, or the results will be skewed. Cover four types:
- Category: "best project management tool for agencies"
- Use case: "how do I track billable hours across clients"
- Comparison: "Competitor A vs Competitor B" and "alternatives to Competitor A"
- Problem: "why are my invoices paid late"
Take wording from Search Console queries, sales call notes, support tickets and Reddit threads. Then add the same competitors you entered in the audit.
type,prompt category,best invoicing software for freelancers in 2026 use_case,how can a small agency send recurring invoices automatically comparison,alternatives to Competitor A for small teams comparison,Competitor A vs Competitor B for freelancers problem,how do I get clients to pay invoices faster
Run every prompt on every engine and log mentions
Run each prompt in a fresh, logged-out or incognito session on each engine you track. For every answer, record whether your brand was mentioned and which competitors were mentioned. Calculate your rate the way the audit does: get your mention percentage per engine, then average those engine percentages.
For the competitor rate, divide the strongest competitor's total mentions by (prompts × engines) and multiply by 100. Subtract the competitor rate from yours to get the score.
Columns: prompt | engine | you (1/0) | comp_a (1/0) | comp_b (1/0) | cited_urls Your rate per engine: =AVERAGEIFS(C:C, B:B, "Perplexity")*100 Your overall rate: =AVERAGE(<one cell per engine rate>) Top competitor rate: =MAX(SUM(D:D), SUM(E:E)) / (COUNTA(UNIQUE(A2:A)) * <number of engines>) * 100 Gap: =<your overall rate> - <top competitor rate>
Isolate gap prompts and the sources behind them
Filter the sheet to rows where a competitor = 1 and you = 0. These are your gap prompts. Use the cited_urls column, or the source panel in Perplexity, ChatGPT search and AI Overviews, to note which pages the engine relied on.
Group the gaps into causes:
- You have no page answering that question.
- You have a page, but it isn't cited.
- The engine cites third-party lists or reviews that leave you out.
Let AI search crawlers reach your pages
Your content can't be cited if retrieval bots are blocked. txt and check that the search and retrieval agents are allowed: OAI-SearchBot, ChatGPT-User, PerplexityBot, Claude-SearchBot and Bingbot (Copilot and some ChatGPT results rely on Bing). Training crawlers such as GPTBot, ClaudeBot and Google-Extended are separate tokens, and you can decide on them independently.
txt allows them.
User-agent: OAI-SearchBot Allow: / User-agent: ChatGPT-User Allow: / User-agent: PerplexityBot Allow: / User-agent: Claude-SearchBot Allow: / User-agent: Bingbot Allow: / User-agent: Googlebot Allow: / Sitemap: https://example.com/sitemap.xml
Publish pages that answer each gap prompt directly
For each cluster of gap prompts, create or rewrite one page that answers it in the first two sentences. Then back the answer up with specifics: prices, limits, integrations, who the product is for and who it isn't for. Formats that get cited most often for competitive prompts are:
- Alternatives pages: /alternatives/competitor-a
- Honest comparison pages with a feature table
- Use-case pages: /invoicing-for-freelancers
Use plain HTML tables and headings that match how the question is phrased. Make sure the key text is in the server-rendered HTML and not loaded only by JavaScript.
<h1>Example Co vs Competitor A: invoicing for freelancers</h1>
<p><strong>Short answer:</strong> Example Co suits freelancers billing under 50 clients a month who need recurring invoices and Stripe payouts; Competitor A suits teams that need built-in payroll.</p>
<table>
<thead><tr><th>Feature</th><th>Example Co</th><th>Competitor A</th></tr></thead>
<tbody>
<tr><td>Starting price</td><td>$9/month</td><td>$15/month</td></tr>
<tr><td>Recurring invoices</td><td>Yes, all plans</td><td>Pro plan only</td></tr>
<tr><td>Payroll</td><td>No</td><td>Yes</td></tr>
</tbody>
</table>
<p>Last updated: 2026-03-01</p>Make your brand a clear, consistent entity
Assistants are more likely to name a brand when they can resolve it confidently. Add Organization JSON-LD to your homepage with your legal name, a one-line description of what you do and for whom, and sameAs links to every official profile. Use that same one-line description on LinkedIn, Crunchbase, G2/Capterra, GitHub and app store listings, so every source repeats the same category wording.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Example Co",
"url": "https://example.com",
"logo": "https://example.com/logo.png",
"description": "Example Co is invoicing software for freelancers and small agencies.",
"sameAs": [
"https://www.linkedin.com/company/example-co",
"https://www.crunchbase.com/organization/example-co",
"https://www.g2.com/products/example-co",
"https://www.wikidata.org/wiki/Q00000000"
]
}
</script>Get included in the third-party sources AI cites
Go back to the cited_urls from step 3. If engines rely on "best X" listicles, review sites or Reddit threads that leave you out, work through them one by one:
- Pitch the authors of those listicles with a one-paragraph summary and a free account they can test.
- Ask existing customers for reviews on the review platforms that get cited.
- Answer relevant community threads openly as a company representative.
Aim to appear on the specific URLs the engines already cite, not on new low-authority sites.
Re-measure with the same prompts and engines
Wait 3–6 weeks after the pages are indexed and placements go live. Then re-run the frozen prompt set on the same engines, using the same run count, and re-run the audit. Compare your averaged rate with the strongest competitor's rate.
You pass when you are no more than 10 points behind. Gap prompts that still fail go back into step 3 for another round.
Platform-specific notes
WordPress
Edit robots.txt in Yoast SEO (Yoast SEO → Tools → File editor) or Rank Math (Rank Math → General Settings → Edit robots.txt). Security plugins such as Wordfence and some hosts block unfamiliar bots, so check their blocklists for PerplexityBot and OAI-SearchBot. Yoast and Rank Math already output Organization schema; add sameAs URLs under each plugin's site representation or social profiles settings.
Shopify
Edit robots.txt by creating a robots.txt.liquid template (Online Store → Themes → Edit code → Add a new template → robots). Build comparison and alternatives pages as standard Pages with a custom template so the tables render in the HTML. Add Organization JSON-LD in theme.liquid or through your theme's structured data settings.
Webflow
Set robots.txt under Site settings → SEO → Indexing. Put JSON-LD in Site settings → Custom code → Head code for the homepage, or in a page's custom code. Use a CMS Collection for alternatives and use-case pages so they all follow one consistent, scalable template.
Next.js
Generate robots.txt with app/robots.ts using the MetadataRoute.Robots rules array, one entry per user agent. Render comparison pages as Server Components or with static generation so AI crawlers, which mostly don't run JavaScript, receive the full text and tables.
Cloudflare
Cloudflare can block AI crawlers by default on newer zones. Check Security → Bots (AI Crawl Control / Block AI bots) and any WAF custom rules. Set retrieval agents such as OAI-SearchBot and PerplexityBot to Allow even if you block training crawlers. Turn off managed robots.txt if it overrides your own rules.
Check that it worked
✅ How to verify
- Re-run the audit and confirm the message shows your mention percentage within 10 points of your strongest competitor.
- Re-run the frozen prompt set on each engine and check that the gap prompts from step 3 now include your brand, ideally with a citation to your new page.
- Check server logs or Cloudflare analytics for requests from OAI-SearchBot, PerplexityBot and ChatGPT-User to your new comparison and use-case URLs.
- Test robots.txt with Google Search Console's robots.txt report and a curl request such as: curl -A "PerplexityBot" -I https://example.com/alternatives/competitor-a (expect 200, not 403).
⚠️ Common mistakes
- Including your brand name in the prompts, which inflates your mention rate and hides the real gaps.
- Changing prompts or engines between measurements, so before and after numbers can't be compared.
- Blocking all AI bots at the CDN while publishing content meant to be cited by those same AI search engines.
- Publishing thin "vs" pages that only praise your product, which assistants rarely cite and which can come across as spam.
Is this still an issue on your site? Run a free audit — ZORO checks this and 294 other SEO, AI-visibility, accessibility and security signals, with fixes for your platform.
Check my site freeFAQ
What is AI share of voice?
It is the percentage of AI assistant answers to category-relevant prompts that mention your brand, compared with the same figure for competitors. Here it is averaged across engines such as ChatGPT, Perplexity, Gemini and Google AI Overviews.
Why does ChatGPT recommend my competitors but not me?
Usually for one of three reasons: you have no page that directly answers the prompt, AI search crawlers are blocked from your site, or the third-party lists and reviews the model relies on don't include you. The citations in each answer show which of these applies.
How long does it take to improve AI visibility?
Search-backed answers in Perplexity, ChatGPT search and AI Overviews can change within a few weeks of new pages being indexed. Answers that come from model knowledge alone may only change after the next model update, which can take months.
Do I need a paid AI visibility tool?
No. A spreadsheet and manual runs work for 30–50 prompts. Tools become worthwhile once you track hundreds of prompts, several markets or weekly trends, because they automate repeat runs and citation logging.
How to Fix Negative AI Sentiment About Your Brand
How to Get Your Brand Cited in Microsoft Copilot Answers
How to Add Citation-Worthy Statistics for AI Answers
How to Get Your Brand Mentioned in Claude Answers
How to Add Concise Answer Blocks for AI Search Engines
How to Get Your Brand Mentioned in Google Gemini Answers