Why Paid Ads Don't Buy You a Citation in ChatGPT

🇫🇷 Version française

On Google, ad spend buys you a guaranteed spot at the top of the page. Many companies assume an equivalent mechanism exists for showing up in a ChatGPT or Perplexity answer. It doesn't — and understanding why changes how you should split an acquisition budget.

Twenty years of paid search have trained companies to expect a simple rule: pay more, get more visibility, instantly. An AI-generated answer works on a different logic. It's built from whatever the model judges to be the most relevant, reliable source for the question asked — not from space sold at auction. Confusing the two leads to unrealistic expectations, and sometimes to badly allocated budgets.

Why the classic paid-search model doesn't transfer

Paid search relies on a limited, sellable slot: a fixed number of sponsored positions, auctioned off, above the organic results. A generative AI answer doesn't have that structure. It doesn't list ten results, some paid and some not — it produces a single piece of text, built from the sources the model judges best suited to the question. There's no equivalent slot to "buy" in that generation process, because the answer itself isn't a sellable space the way a results page is.

What paid advertising still does very well

That doesn't make ads useless — they simply serve a different goal. They're still effective for driving immediate traffic to a specific offer, quickly testing messages and audiences, or supporting a launch while organic content is still being built. The key difference is how long the effect lasts: ad-driven visibility stops the moment the budget stops, while a presence built in AI answers keeps existing without ongoing spend, as long as the content behind it stays relevant and up to date.

The indirect effect of paid traffic on GEO

Ad spend doesn't directly buy a citation, but it can contribute indirectly. Well-converted paid traffic generates customers, which can in turn generate reviews, testimonials, mentions on forums or social platforms — third-party content an AI can later use as proof of reliability. That's a real side effect, but a slow and indirect one: it doesn't replace the underlying work on the brand's own content, which remains the deciding factor.

What this means for budget allocation

The practical takeaway isn't to pit the two levers against each other, but to give each a different role: advertising for speed and testing, content and site structure for durable presence in AI answers. A company that pours its entire budget into ads hoping it also "feeds" its AI visibility is making a bet that doesn't hold up — that's a separate editorial and technical effort, built over time.