AEO can be worth paying for when the questions you target relate to a valuable purchase and your business can turn the resulting attention into customers. The decision gets easier once you put a number on what that customer is worth.

Before hiring an AEO agency, work out what a useful engagement would need to achieve for your business. You do not need a prediction about the future of search to do that calculation.

You need a realistic estimate of the contribution a customer makes after the costs of serving them, and a clear idea of what you’re paying to acquire one. An impressive screenshot from ChatGPT doesn’t answer either question.

Start with a budget you can explain

Consider a hypothetical consultancy evaluating a $3,000 monthly engagement. That figure is an example for the calculation, not an AEO market average or our price.

Suppose a new client contributes $6,000 after delivery costs over the period the consultancy uses to evaluate acquisition. Two months of agency fees would require one additional client to cover those fees. The consultancy still has to account for the time its own staff spend supporting the work.

Now change the business. A store selling a product with $30 of contribution per order would need 100 additional orders to cover the same monthly fee. Returns and repeat purchases could change that calculation. The store needs its own numbers.

The visibility opportunity could look similar in a slide deck. The spending decision is different because the underlying businesses are different.

Use a payback period your finance team accepts. Don’t quietly switch from first-order profit to an optimistic lifetime value because the second number makes the proposal easier to approve.

Break-even customers = total project cost ÷ contribution per additional customer. Use the same evaluation period for both figures, include your internal delivery effort in project costs, and round up when planning whole customers. Our guide to AEO pricing explains which costs to include when comparing proposals.

The questions need to attract possible customers

A company selling bookkeeping services can answer “what is accounting?” It can also answer a question about replacing a part-time bookkeeper when a business grows.

Both questions belong to the same subject. The second gives the company a much clearer reason to participate.

Before funding a campaign, inspect the proposed question set. For each important question, describe the person asking and the next decision they might make. If that person is unlikely to need what you sell, give the question less weight.

This doesn’t mean every article must demand a purchase. A useful explanation can build trust before a buyer is ready. You should still be able to explain why that reader matters to the business.

Fix the obvious problems before buying a large program

If your service page doesn’t explain what is included, start there. If nobody follows up on inquiries, solve that too.

Sending more people into a confusing offer makes the numbers harder to interpret. You might conclude the channel doesn’t work when the problem is the page or the sales process.

A focused first scope is easier to evaluate. Choose a commercially important part of the site and agree on the work required. Then give the resulting pages time to be discovered and observed, while keeping a record of the changes.

The first review date is a management decision. It isn’t a promise that an AI system will recommend you by then. A focused AEO assessment and implementation scope should identify the pages, deliverables and responsibilities before you commit.

Count evidence without giving it more credit than it deserves

OpenAI says ChatGPT referral URLs include utm_source=chatgpt.com, which can help identify arriving visitors in analytics. That is useful evidence when the visitor follows a trackable link. OpenAI’s publisher FAQ describes the parameter.

Other buyers may see your name and look you up later. Ask new inquiries how they found you, while recognizing that their recollection won’t be perfect. Keep that answer alongside the tracked source rather than rewriting the attribution to suit the campaign.

You may finish an initial test with better evidence of visibility and too few sales to judge the economics. Say that plainly. A small sample is a reason to remain uncertain, not permission to multiply an encouraging percentage into an annual forecast.

Know what would make you stop

Before extending an engagement, look at whether the work is producing commercially relevant progress and whether there is a credible next action.

If the agency keeps finding you in answers about subjects outside your offer, refine the scope. If meaningful questions remain unaddressed because your team can’t supply the necessary facts, solve that bottleneck before ordering more content.

Use the observations above at the next budget review. Our guide to what an AEO agency should deliver helps you check the responsibilities and outputs in a proposal. For a concrete example of the work, see how a buyer question becomes useful AEO content. If you cannot explain what another month would help you learn or improve, ask for a smaller next step before renewing.