Cheap AEO packages fail because they spend the money on the wrong side of the fence. In 2026 a working AEO retainer runs $2,500 to $10,000 a month, and packages priced under $1,000 mostly buy on-site work: schema markup, a few FAQ blocks, meta rewrites. Ahrefs tracked 1,885 pages that added JSON-LD schema between August 2025 and March 2026 against 4,000 matched controls and measured no meaningful change in citations from Google AI Overviews, Google AI Mode, or ChatGPT. The engines build answers from third-party sources, and a sub $1,000 package cannot touch those.
What does a cheap AEO package actually include?
Almost always the same four items, and all four live on your own website. Schema markup across service pages. An FAQ section appended to existing content. Meta title and description rewrites. A monthly visibility report from a tool like Otterly.ai or Semrush’s AI Visibility Toolkit, which the vendor buys for $29 to $99 a month and resells inside the package.
None of that is worthless. Schema helps rich result eligibility under Google’s structured data guidelines. FAQs give a page a clearer question-and-answer shape. Tracking tells you where you stand. The problem is that the package is sold as a way to get recommended by ChatGPT, Perplexity, and Google AI Mode, and the mechanism by which those engines choose a business does not run through your meta descriptions.
Look at where citations come from. Semrush’s three-month study of AI citations put Reddit and Wikipedia at the top of the source list across engines, followed by established publishers, directories, and review platforms. Ahrefs’ analysis of 17 million AI citations found the cited URLs were on average 25.7 percent fresher than organic search results, with ChatGPT skewing roughly 458 days newer. The engines are reading recent, third-party pages about you. A cheap package rewrites your own pages and waits.
Why do the engines ignore what cheap packages optimize?
Because the highest-value signals are off-site, and the only controlled test of the main on-site lever came back flat. Three findings decide this.
First, the schema test. Ahrefs’ study of 1,885 pages found AI Overview citations down 4.6 percent relative to controls after schema was added, with AI Mode up 2.4 percent and ChatGPT up 2.2 percent, both inside the noise. Every page in that dataset already had 100 or more AI Overview citations before schema went on. If schema moved citations, that is the sample where it would have shown. It did not. Schema remains worth implementing for rich results and machine readability. It is not a citation lever, and a package built around it is selling the wrong thing.
Second, the click data. The Pew Research Center followed 900 U.S. adults through March 2025 and found users clicked a traditional link on 8 percent of Google visits that showed an AI summary, against 15 percent without one, and clicked a source inside the summary on just 1 percent of visits. Ahrefs measured a 34.5 percent lower clickthrough rate for the top result across 300,000 keywords when an AI Overview appeared. The visitor who does arrive is valuable, which is why Semrush’s 2025 conversion research found AI search visitors converting at roughly 4.4 times the rate of standard organic traffic. But you cannot buy that visitor with an FAQ block.
Third, the source mix. ChatGPT search rewrites your prompt into targeted queries and sends them to third-party search providers including Microsoft Bing plus licensed publisher content, according to OpenAI’s own documentation. If Bing has not indexed your site, or the pages that answer the buyer’s question about your category live on Reddit, Avvo, RealSelf, Healthgrades, or a trade publication, the engine never needs your homepage to build its answer.
Curious what the engines actually say about your business right now? Get a free AI visibility check and see the gap before you buy a package to close it.
Cheap package vs full retainer vs doing it yourself: which fits?
Most businesses under about $2,500 a month in total marketing budget should not buy any AEO service. Above that line, the question is scope. Here is the comparison buyers are actually making.
| Option | Monthly cost | What it covers | What it cannot do | Best for |
|---|---|---|---|---|
| Cheap AEO package | Under $1,000 | Schema, FAQ blocks, meta rewrites, a resold tracking report | Third-party mentions, reviews, press, directory presence, Bing indexing fixes | Nobody as a growth play. Acceptable as a one-time technical cleanup if priced as a project |
| Tracking tool only | $29 to $250 | Prompt-level monitoring across ChatGPT, Perplexity, Google AI Mode, Claude | Anything that changes the answer | Businesses under $2,500 a month in marketing budget who want to measure before spending |
| Mid-tier AEO retainer | $2,500 to $5,000 | Content built around buyer questions, review generation, directory and citation cleanup, monthly tracking | Placements in national press, large-scale digital PR | Single-location professional practices and small firms with a defined niche |
| Full-scope AEO plus PR retainer | $5,000 to $10,000 | Everything above plus earned and placed media, reputation work, multi-engine tracking | Guarantee a specific citation on a specific prompt | Firms above roughly $1M revenue in competitive metros |
| Do it yourself | $0 plus 6 to 10 hours a week | Claiming and completing every profile, answering journalist queries, publishing question-format content | Scale, media relationships, speed | Owners with writing time and one clear area of expertise |
| Do nothing | $0 | Whatever third parties already say about you | Change the answer | Businesses at capacity with no acquisition need |
The cheap package row is the one buyers argue with. The objection is reasonable: it is a low-risk test. The problem is that it tests the wrong hypothesis. Spending $700 a month on schema and FAQs for six months tells you whether schema and FAQs produce citations, and Ahrefs already ran that test with 1,885 pages and a control group. What it does not tell you is whether your business is citable at all, because the levers that decide that were never pulled.
If you want a cheap test, buy the tracking tool for 90 days and spend the rest of the budget on reviews and one press placement. That tests the real mechanism for less than the package costs.
What separates a package that works from one that does not?
Whether the money goes off-site. A retainer that earns its fee spends most of the hours on things you do not control directly: reviews on Google Business Profile and the platform that governs your vertical, complete and consistent listings on the directories the engines actually cite, and fresh third-party pages from press or contributed articles. The rest goes into content shaped around the questions buyers type into ChatGPT, which is a different job from rewriting existing service pages.
There are specific things to look for in a proposal. Named platforms rather than “AI optimization.” A plan for Bing indexing, since ChatGPT search draws on it. A review cadence with a target count. A press or placement component, even a small one. Reporting that tracks named prompts across named engines and shows which sources the engine cited, not just whether your brand appeared.
And there are specific claims that should end the conversation. Any vendor asserting that structured data is what gets you cited. Any guarantee of a citation on a specific prompt. Any package that lists “AI submission” or “LLM registration” as a deliverable, since no such submission process exists at OpenAI, Anthropic, Perplexity, or Google. Our list of AEO agency red flags covers the rest.
Setup fees deserve one sentence: a fair one is $0 to one month of the retainer, itemized. The reasoning is in our guide to what a fair AEO setup fee looks like.
How should a buyer with a limited budget approach AEO?
In the order the engines weigh things, not the order vendors sell them. Here is a sequence that fits a business with $1,500 to $3,000 a month to spend.
1. Measure first, for 90 days
Buy a tracking tool at $29 to $99 a month. Pick 20 prompts a real buyer would type. Record which businesses and which sources the engines name today. This baseline is what every later dollar gets judged against.
2. Fix the off-site record
Claim and complete every profile that matters in your category: Google Business Profile, Bing Places, and the vertical platform, whether that is Avvo and Martindale-Hubbell for a law firm or RealSelf and Healthgrades for a surgeon. Match name, address, and phone everywhere. This costs staff time, not vendor fees.
3. Build review volume on the platforms the engines read
BrightLocal’s 2025 Local Consumer Review Survey found 83 percent of consumers read reviews on Google, and the same profiles feed local answers in AI engines. Set a monthly target and ask every satisfied client.
4. Publish answers, not pages
Write to the prompts from step one. One question per page, the answer in the first two sentences, a number in the first paragraph. This is the part of a cheap package that is worth keeping, done properly.
5. Add third-party pages as budget allows
One placement a quarter in a publication the engines already cite does more than a year of meta rewrites. If you reach the point where a retainer makes sense, the service tiers are priced by scope rather than by page count, and the entry tier starts at the point where off-site work is included.
Frequently asked questions
Are cheap AEO packages ever worth buying?
As a one-time technical project, occasionally. If your site has no structured data, no FAQ content, and broken meta tags, a $500 to $1,500 fixed-price cleanup is reasonable. As a monthly growth service under $1,000, no. The recurring fee pays for reporting on work that does not move citations, and the money is better spent on reviews, listings, and one press placement.
What is the minimum monthly budget for AEO that works?
Roughly $2,500 a month including tools, and that assumes a single-location business with a defined niche. Below that, buy a tracking tool, do the profile and review work in-house, and hold the rest. Above it, a retainer can fund off-site work: review generation, directory cleanup, question-format content, and at least some earned media.
Does schema markup help with AI citations?
The only controlled test on record says no. Ahrefs tracked 1,885 pages that added JSON-LD schema against 4,000 matched control pages and found AI Overview citations down 4.6 percent, with AI Mode and ChatGPT changes too small to distinguish from noise. Schema is worth implementing for rich result eligibility and machine readability. It should not be sold as the reason you will be cited.
Why do AI engines cite Reddit and Wikipedia so much?
Because those pages are fresh, specific, and answer questions in the language people ask them. Semrush’s three-month citation study found both at the top of the source list. Ahrefs’ 17 million citation analysis found cited URLs 25.7 percent fresher than organic results on average. Neither site is something you can buy your way onto, which is exactly why a cheap on-site package cannot replicate the signal.
How long does real AEO take to show results?
Expect 60 to 90 days before tracked prompts start moving, and six months before the change is stable across engines. Google’s own guidance for hiring an SEO says four months to a year to see benefit, and AI engines add a freshness lag on top. Any package promising citations in 30 days is either pitching a prompt nobody types or counting a mention that was already there.
Can I do AEO myself instead of buying a package?
Yes, and for businesses under roughly $1M in revenue it is often the right call. The work is claiming and completing profiles, generating reviews, answering journalist queries, and writing question-format content. It costs six to ten hours a week rather than a retainer. What you cannot easily do alone is earn press at scale, which is where a retainer eventually earns its fee.
The price is not the problem, the scope is
A cheap AEO package fails for a structural reason: everything it delivers sits on your own website, and the engines build their answers from everyone else’s. Ahrefs tested the main on-site lever against a control group and found nothing. Semrush and Ahrefs both show the citations coming from fresh third-party pages. Pew shows the clicks shrinking, which makes the answer itself the thing you are competing for. Spend the small budget on measurement, reviews, and listings. Spend the larger budget, when you have it, on the off-site work that a $700 package was never going to include.
Want the baseline before you decide? Request a free audit and get a read on which prompts name you today and which sources the engines are using instead.
Tagged