Perplexity Ads: Current Status and How to Get Cited

Perplexity has pulled back from advertising, so do not build your next campaign around buying Perplexity ads. The Financial Times reported the retreat on 18 February 2026, as Search Engine Land covered that day; the practical route to visibility in Perplexity's answers is to make your business a useful, verifiable source. For a paid placement in an AI conversation, ChatGPT has a documented self-service buying route.

Status checked against Perplexity's business pages and Help Center, and OpenAI's advertiser documentation, on 22 October 2026. The withdrawal is reported news; the distinction between Perplexity's research product and an ad-buying service is visible on its own pages.

Can You Buy Perplexity Ads Today?

There is no current self-service buying route verified by the official Perplexity pages reviewed for this article. For a marketer planning a launch, that means keeping Perplexity out of the committed paid-media plan unless the company supplies a current, explicit offer. A surviving announcement about an experiment is insufficient evidence that inventory is available.

Perplexity, an AI search service that answers questions using web sources, currently has a page called Enterprise for Advertising. It describes research for campaign planning, insights from the public web, and incorporating internal briefs or brand guidelines. Its Get started button leads to organization setup. It does not present a workflow for buying placements inside Perplexity.

The current Help Center likewise documents the product, subscriptions, enterprise features and API. Its English documentation index did not provide an advertiser-buying guide when checked. That is a bounded public-page finding, not a claim that every private commercial agreement is knowable.

Perplexity AI Ads: What the Original Experiment Offered

Perplexity's original advertising announcement described sponsored follow-up questions and paid media beside an answer. It said sponsored material would be labeled and that advertisers would not write or edit the answers to sponsored questions.

Those are historical formats, not a current menu. The same page remaining online does not establish that you can open an account, reserve a placement or launch a campaign today.

Perplexity Marketing: The Research Use Case

An agency can use the enterprise research product to investigate a client's market. That is a different purchase from paying for the client's brand to appear in someone else's answer.

For a local-services founder, the distinction is practical: paying for better research access does not buy a recommendation when a customer asks for a provider. Keep the subscription decision in your software budget and the visibility work in your marketing plan. The Perplexity pricing guide covers the product plans separately.

What Changed, and Where the Reports Disagree

The reporting supports a retreat from advertising, but it does not support one uncontested shutdown date or an irrevocable ban. The buyer's decision is clearer than the timeline: a new campaign needs a confirmed current buying route.

The reported concern was that advertising could make users question the independence of answers. That explains the retreat; it does not establish that every sponsored placement influenced an answer, or that every AI advertising product operates the same way.

Treat Perplexity as an earned-visibility project for planning purposes. Revisit the paid-media decision when current first-party buying documentation or a specific commercial offer changes the evidence.

What Gets a Business Into Perplexity's Answers?

Your business needs useful evidence that Perplexity can find and choose to cite. A citation is the link supporting an answer, much like a footnote in a research memo. The memo may discuss your company while citing somebody else's reporting about it.

Perplexity's own explanation says it searches the web and provides answers with citations and links to original sources. That gives you a concrete objective: make the facts a prospective customer needs available on relevant, verifiable pages.

For a business-to-business software company, useful candidates include integration documentation, migration instructions and a clear account of product limitations. For a retailer, they include product specifications, compatibility, delivery conditions and returns. For a local-services business, they include service areas, qualifications and the conditions that determine whether a job is suitable.

These are publishing recommendations, not disclosed Perplexity ranking factors. A page deserves attention when it resolves a real buying question, even before any AI system cites it.

Independent evidence matters for a different reason: your company controls its own claims, but a buyer may want corroboration. Maintain accurate public business information and provide useful documentation to people independently reviewing your product. If relevant customer discussions are cited, understand the concern being discussed. Do not turn that observation into a plan to manufacture reviews or flood forums.

One Checklist for Getting Cited

Start with a buyer question and finish with an evidence log. This checklist is a work plan, not a promise of placement or a timetable for being indexed.

1. 1

   1. Choose questions connected to a sale

   Use questions your customers ask before choosing you. A software growth lead might start with integration compatibility and migration effort; a service owner might start with coverage and job eligibility. Save the exact wording so later checks are comparable.

2. 2

   2. Assign each question a public source page

   Identify the page that should answer it. Put essential factual information on an accessible product, service or documentation page. A private sales deck may help your sales team, but it gives a public answer engine little to cite about that claim.

3. 3

   3. Check PerplexityBot access with the site owner

   Perplexity's crawler documentation recommends allowing PerplexityBot, its search crawler, in robots.txt, the file containing your site's crawler instructions, and permitting its published IP ranges. Ask the technical owner to check firewall rules as well as crawler instructions, using the documented user-agent and IP conditions. Keep private customer and account pages protected.

4. 4

   4. Check the text the public page delivers

   Make sure the answer is available without signing in or completing a form. Have the site owner inspect what a crawler receives if key content depends on scripts or interactions. Perplexity's robots guidance says a blocked page may still contribute a domain, headline or brief summary, even when its full text is not indexed. Seeing your domain is therefore not proof that the detailed evidence is readable.

5. 5

   5. Answer directly and preserve the conditions

   Name the product or service, give the answer, then state the limitation. A retailer's compatibility page should specify what fits and what does not. A software page should say which integration actions are supported. Remove broad claims that the underlying documentation cannot support.

6. 6

   6. Attach evidence and an accountable owner

   Link to the relevant specifications, policy or documented method beside the claim. Use original measurements only when you have them. Assign a person who knows the product or service to approve the facts and update them when something changes.

7. 7

   7. Resolve inconsistent business information

   Check whether your product page, help article and relevant external business listing describe the same service accurately. Correct obsolete information where you can. Consistency helps a buyer verify the business; it is not a guaranteed ranking boost.

8. 8

   8. Recheck the same questions and save the sources

   Record the answer, the exact cited URLs and whether the linked pages support the claims. Track qualified inquiries separately. Expand the work when it improves useful customer information, not merely because a dashboard score moves.

For the page-editing method, use answer engine optimization. For a scheduled follow-up, the 14-day GEO playbook provides a broader implementation structure. GEO means improving visibility in generative answers; its schedule is a work schedule, not a citation guarantee.

How to Check Whether You Are Cited

Open the sources and inspect the actual linked page. Asking an AI whether it knows your business is not enough to establish that your website supplied an answer.

How to Use Perplexity AI for Research

Begin with a fresh conversation and an ordinary customer question that does not name your brand. For a software company, an illustrative prompt is: “Which appointment tools support exporting customer records, and what does each export include?” For a service business, use the genuine service requirement and location.

Save the date, exact prompt, relevant search settings, answer and source links. Open each citation associated with your business or the feature you care about. Check the destination domain, exact page and supporting passage. A homepage link and a link to the specific documentation you improved are different observations.

Then run a branded question as a separate diagnostic. Supplying your company name or URL can help you inspect how it is described, but it changes the task. Keep that result separate from the unprompted discovery check.

Keep failed checks and answers with no relevant citation in the log. Repeat the same questions under recorded conditions before treating a change as a trend. This is a sample of observed answers, not a census of every customer's experience.

A founder should also review identifiable Perplexity referrals and the resulting qualified leads or purchases in their analytics and sales records. A citation is not a click, and a click is not a customer. Where the referral cannot be identified, leave it unknown instead of assigning it to AI by guesswork.

Where Paid AI Ads Exist: ChatGPT Self-Serve

ChatGPT, OpenAI's conversational assistant, has a documented route for buying ads through Ads Manager. OpenAI announced the self-service beta on 5 May 2026. Its current availability table lists both the United States and Canada as available and ties eligibility to the advertising and billing entity's country.

The official starting point is OpenAI Ads. Ads Manager provides campaign setup and reporting; review, account readiness and billing still affect whether ads can serve. OpenAI describes the platform as beta and says it does not yet have established performance benchmarks across advertisers, industries or campaign types. OpenAI's advertiser FAQ

For a retailer, the test should connect a specific product need to a useful landing page and a measurable purchase. For a service business, define a qualified inquiry before judging the channel by traffic. The ChatGPT self-serve ads guide covers the buying workflow and measurement in detail.

There is a firm boundary: OpenAI says ads are labeled and separate from responses, and advertisers cannot shape, rank or alter ChatGPT's answers. Ads may appear for Free and Go users; Plus, Pro, Business, Enterprise and Edu accounts do not have ads. A software founder should therefore check audience fit rather than assume the whole ChatGPT user base is reachable. OpenAI's ads explanation

What's Overhyped, and Who Should Act Now?

The misleading promise is that a payment or a formatting trick can secure an independent recommendation. A Perplexity subscription buys product access. A ChatGPT campaign buys the opportunity to serve a separate ad. Source improvements make your evidence more useful and accessible. None of those is a documented purchase of Perplexity's answer.

Do not turn the reported advertising retreat into a claim that Perplexity will never sell ads again. Equally, do not treat the old experiment page as proof that a new advertiser can buy now. Both claims go beyond the current evidence.

Paid-media buyers should act on a confirmed buying route. If ChatGPT fits the audience and the business can measure qualified outcomes, evaluate a bounded test through the existing guide. Wait if measurement or the destination page is not ready.

Content and growth leads should act on missing evidence. A software company whose documentation cannot answer common migration questions has a concrete publishing job. A local operator with conflicting service-area information has a correction job. Both improvements help customers regardless of citation outcomes.

Businesses with accurate, accessible pages do not need wholesale rewrites. Monitor the questions that matter and change facts when the underlying service changes. A fluctuating answer is a reason to investigate, not an instruction to republish everything.

Your Monday move: give one marketer a real buyer question, one page owner the supporting facts, and one technical owner the crawler-access check. Save the baseline answer and its citations before editing. That produces work you can review and evidence you can compare.

Frequently Asked Questions

Does Perplexity have ads?

Reporting describes a withdrawal from its advertising experiment, with disagreement over the wind-down timeline. As of 22 October 2026, the official business and help pages checked here did not establish a current self-service buying route. Do not assume the historical experiment is available to new advertisers.

Perplexity Ads manager

No current Perplexity self-service ads manager was verified in the official pages reviewed. The Enterprise for Advertising page is about using Perplexity for research and campaign planning. ChatGPT's Ads Manager is a separate OpenAI product.

Perplexity ads examples

Perplexity's historical announcement shows sponsored follow-up questions and paid media beside answers. Treat those examples as documentation of the experiment, not proof of current inventory.

Is ChatGPT better than Perplexity?

For the specific job of buying a paid AI placement, ChatGPT has the documented self-service route described above. That does not settle which product is better for research, and buying its ads does not purchase influence over its answers.

Why is Perplexity so popular?

Its source-linked answers make it useful for researching a question and opening the supporting material. For a marketer, that appeal is a reason to maintain useful source pages, not evidence that visibility will produce customers.

For practical guidance on AI and marketing decisions, join the newsletter.