Why Perplexity cites your competitors but not you

You type the question into Perplexity. The one your product is built to answer. It writes a tidy paragraph, lists four sources underneath, and three of them are your competitors. Your page is better. It is nowhere.

That stings in a specific way, because Perplexity shows its work. Google's AI Overview at least hides who it picked, which makes getting into Google's AI Overviews a quieter, harder thing to diagnose. Perplexity names the winners right under the answer, every time, and your name is not on the list. The good news buried in that frustration: the reasons are knowable, and most of them are fixable this week.

What Perplexity is actually doing

Perplexity does not answer from memory. It runs a live search on your question, pulls back a set of candidate pages, reads them, and writes a short answer grounded in what it just read. Then it cites the handful of pages it leaned on hardest. This is retrieval-augmented generation: retrieve first, generate second, cite the sources behind the generation.

Two things follow from that. First, the citations reflect the live web, not a frozen training snapshot, so work you do now can show up in days rather than waiting on a model release. Second, the whole game is retrieval. If your page was not in the set Perplexity pulled back, no amount of quality on the page matters. It was never in the room.

So the question "why them and not me" splits into a sharper one: were you even retrieved, and if so, was your answer easy to lift?

Why them and not you

When competitors get cited and you don't, they are simply more retrievable for that exact query. Four things decide it, roughly in order of how often they're the culprit.

• You're not indexed or ranking for the query. Retrieval starts from a search step. If your page isn't in the index, or doesn't surface in the top results for that phrasing, it never enters the candidate set. This is the same logic as classic SEO, and it's where most "I'm invisible" cases actually live.
• You block the crawler. PerplexityBot indexes pages so they can be retrieved later. If your robots.txt disallows it, often a leftover from a "block all AI" rule someone added in a panic, you've opted out of being a candidate at all.
• Your answer is buried. Perplexity is looking for a clean, self-contained statement it can quote and attribute. If your page circles the topic for five paragraphs before getting to the point, the model quietly takes the cleaner sentence from a competitor who led with it.
• You're not a credible source for the topic. When Perplexity cites you, it stakes a little of its own credibility on you. Thin, one-off pages lose to sites with real depth on the subject. This is the slow one, and it's why an established competitor keeps winning even when your single page is sharper.

None of this is a mysterious AI ranking factor. It's ordinary search retrieval plus one new test: is the answer cleanly extractable? That framing is the heart of answer engine optimization, and the same four culprits explain why ChatGPT isn't citing your website when you run the check there instead.

Reverse-engineer the citations

You don't have to guess. Perplexity hands you the diagnostic. Run these in order.

Type the exact query and study the sources. Use the same phrasing a real customer would. Note which three to five pages it cites. Open each one. What does it do that yours doesn't? Almost always you'll see the same pattern: a plain heading that matches the question, then a direct answer in the first sentence or two. Not better prose. A more liftable answer.

Check whether your page is even in the candidate set. Search the same query on Google and look for your page in the organic results. If you're not ranking there, you're almost certainly not being retrieved by Perplexity either, because both start from a search step. That tells you the problem is retrieval, not extraction, and it changes which fix comes first.

Check that you allow PerplexityBot. Open yoursite.com/robots.txt and look for any rule blocking PerplexityBot. A single stray Disallow line under that user-agent removes you from the running. While you're there, confirm your key page is actually indexed by Google with a quick site:yoursite.com/your-page search.

Check whether your answer is buried. Open your own page and find the sentence that directly answers the query. If it's standalone and near the top, good. If you have to scroll past background and qualifiers to reach it, or if it only makes sense in the surrounding paragraph, that's why a competitor's tighter sentence got quoted instead.

By the end you'll know which of the four causes is yours. Usually it's one or two, not all four.

The fixes, in order

Do these in sequence. Extractable prose is wasted effort if you were never retrievable in the first place.

1. Be retrievable. Get the page indexed, allow PerplexityBot, and earn enough relevance to rank for the query in ordinary search. There is no AEO shortcut around this. If you don't surface in the top results, you don't get pulled into the candidate set, full stop.

2. Write the answer as a clean, self-contained passage. Find the obvious question the page answers and put a direct answer right under a question-shaped heading.

• Lead with the answer. A heading like ## How long does setup take? followed by "Setup takes about ten minutes for a standard store" can be quoted as-is. A paragraph that builds to the point over four sentences cannot.
• Make sentences stand alone. "It usually takes two days" is useless out of context. "Bank transfers clear in about two business days" survives being lifted onto a results page with nothing around it. Write as if every sentence might be quoted naked.
• Keep the real text in the HTML. If the key fact lives inside an image, a PDF or a script that runs after load, the crawler may never read it. If you can't select it with your cursor, assume Perplexity can't either.

3. Build genuine topical depth. One good page rarely makes you the credible source for a whole subject. Cover the cluster: the adjacent questions, the comparisons, the how and the why, with a named author and real expertise behind it. Earn mentions on sites Perplexity already trusts, so when it weighs whether to cite you, there's corroboration to lean on. This is the slow work, and it's the part competitors can't copy overnight.

The robots and extraction fixes can land this week. Ranking and depth take longer, but they're the moat. For the cross-engine version of this playbook, getting cited by ChatGPT and Perplexity goes wider than this one does.

Track who's getting cited

Running this once, by hand, for one query is doable. Watching it across the queries you care about, and re-checking every time you ship a change, is the part nobody keeps up with. Rankport's AI visibility checker shows which queries cite you versus your competitors, reads your robots.txt for each bot, and flags where your answer is too buried to lift, in plain English. If Perplexity is naming rivals instead of you, it tells you which of the four reasons is to blame.