Why “Ranking” Is the Wrong Word for Perplexity

“Ranking” is the wrong word for what happens in Perplexity, and the mismatch causes most of the wasted effort.

Perplexity answers a question and cites the sources it drew on. There is no ordered list of ten results, no position to occupy. The unit of success is being one of the handful of cited sources and critically, those sources are frequently not the websites of the companies being recommended.

That last point reorders the work.

The Three Gates

Three things have to be true, in sequence, and failing an early one makes the later ones irrelevant.

1. PerplexityBot Can Reach You

Check yoursite.com/robots.txt for PerplexityBot and Perplexity-User. If either is disallowed, nothing else applies.

2. It Receives Actual Content

Perplexity's crawler, like most AI crawlers, does not execute JavaScript. If your site is built with React, Vue, Angular or similar and renders in the browser, the crawler may receive an empty shell.

Check with:

curl -A "PerplexityBot" yoursite.com | wc -w

Under fifty words on a site with real content means the crawler is getting nothing. Prerendering or server-side rendering fixes it, and no content work matters until it's done.

3. Something Worth Citing Exists

Access is necessary, not sufficient.

Gates one and two are engineering. Most sites fail at least one and never find out, because there is no report and no alert only silence that looks identical to not being good enough.

The Part That Differs From Google

Here is the structural difference worth planning around.

For a question like “what's the best marketing tool for a small business,” Perplexity typically cites roundups, comparison articles, review sites and forum discussions — not vendor homepages. A vendor's own page is a weak source for a recommendation question, and the system behaves as though it knows that.

So the highest-leverage work is often not on your website at all. It's getting into the pages that get cited.

Practically:

Find the Cited Sources

Ask Perplexity the questions your buyers would ask. Record every source it cites, across a dozen or so query variations. Patterns emerge quickly — the same handful of articles and sites recur.

Get Included in Them

Some are open to submissions. Some are journalist-written roundups where a pitch is possible. Some are review platforms where you can create a profile. Some are communities where participation is the route in.

This is closer to digital PR than to SEO, and it's the part most “GEO” advice skips because it can't be automated.

Review Platforms Are Disproportionately Useful

G2, Capterra, GetApp and Software Advice are heavily cited across recommendation queries. G2 acquired the other three in February 2026 but kept their review pools separate — a profile on one does not create presence on the others.

Capterra listings are free and syndicate across several sites, which makes it the highest-return starting point for most businesses.

What to Do on Your Own Site

Your site still matters for the queries where someone is asking about you specifically, and as corroboration when another source names you.

Answer the Question Near the Top

AI systems extract passages, not pages. A direct answer in the first paragraph is more likely to be quoted accurately than one that arrives after four paragraphs of context.

Be Specific and Checkable

Prices, numbers, named constraints, clear statements of what you do and don't do. Vague copy is hard to cite because there's nothing in it to quote.

Use FAQ Markup

Question-and-answer structure with FAQPage structured data maps well onto how these systems retrieve.

Publish Dates

Use datePublished and dateModified in structured data. Unknown recency reduces confidence.

Comparison Pages, Honestly Done

Comparison content ranks and gets cited for recommendation queries. Accuracy is the binding constraint — competitor pricing changes without notice, and a wrong claim is worse than no page.

What Doesn't Help

Two things sold as Perplexity optimisation that don't survive scrutiny.

llms.txt Files

Google states directly that Google Search doesn't use them. Perplexity has published nothing committing to them either. Treat as optional, never foundational.

Keyword-Stuffed Rewriting for AI

These systems work on meaning rather than phrase matching. Rewriting content to hit variations produces worse content for humans and no measurable advantage.

Measuring

There is no Perplexity equivalent of Search Console. The workable substitute is a manual log, run monthly:

Twenty queries takes about half an hour. The third column is the valuable one — it produces a concrete list of pages to get into, which is more actionable than knowing you're absent.

Run the same queries monthly. Changes there often precede changes anywhere else.

Realistic Expectations

Fixing crawler access has an effect within weeks, because it changes what's in the index.

Getting into third-party sources takes months, because it depends on other people publishing. It is also the more durable outcome: a citation in a widely-referenced roundup keeps working, and it isn't something a competitor can undo by changing their own site.

The uncomfortable summary is that Perplexity visibility is substantially outside your direct control, and the parts inside your control are mostly technical. Effort spent rewriting homepage copy for AI is effort not spent on the two things that move it.

Questions

Is Perplexity Worth Optimising for Given Its Size?

The volume is small relative to Google. The intent is not — Perplexity users are frequently in a research or comparison mindset, and being the cited recommendation carries more weight than a position in a list. Worth the work if your buyers research before purchasing.

Does Perplexity Use Google's Index?

It maintains its own crawling infrastructure and also draws on other sources. Its crawler needs its own permission regardless of your Google standing.

Can I Pay to Appear?

Perplexity has run advertising products. Citations in answers are not purchasable, and any service claiming to guarantee them is claiming something it can't control.

How Is This Different From Optimising for ChatGPT?

ChatGPT search draws on Bing's index, so Bing Webmaster Tools matters there in a way it doesn't for Perplexity. Perplexity leans more heavily on third-party sources and forums. The technical requirements — crawler access, real HTML — are identical for both.

Do Backlinks Matter?

Indirectly. A link from a source Perplexity trusts brings both the reference and the traffic. But being mentioned in a cited source matters even without a link, which is a different target from classic link building.