Standalone service or SEO extension? How to package AI visibility in your agency
You've already decided to offer AI visibility — getting the AI to mention and recommend your clients when someone asks it — and now comes the decision almost nobody stops to think about: do you put it inside the SEO retainer or sell it as a service with its own name?
It looks like a packaging detail. It isn't. That decision determines how much you can charge, who you can sell it to, what happens to your retention and how you position yourself against the agency next door. And the right answer isn't the same for an SEO agency with a roster of retainers as it is for a web design studio that lives off projects.
Let's go through the two models, their numbers and the recommendation for your case.
Model 1: SEO extension
GEO (Generative Engine Optimisation: working a brand's presence in the answers from ChatGPT, Gemini, Perplexity and company) comes in as an expansion of the current retainer. Same contract, same invoice, one more line.
What it has going for it:
Less sales friction. The client already pays you every month and already trusts you. There's no new proposal, no buying process, no comparing providers: it's "we add this to the service for X € more". For an existing roster, it's the shortest sale you'll make this year. The exact email and meeting for that conversation are in how to sell AI visibility to your SEO clients.
Retention. Every additional service a client takes on reduces the chance they'll leave: cancelling a retainer of "SEO + AI visibility" hurts more than cancelling plain SEO. And there's a huge defensive effect: if your client hears about GEO and you don't offer it, someone else will — and that other agency will, while they're at it, sow doubts about your SEO.
Coherent narrative. When organic traffic drops because of AI Overviews — the average organic CTR drop in Spain is around 61%, according to ismajimenez.com — the extension lets you tell a single story: "search is changing and your service changes with it", instead of defending SEO on one side and selling something else on the other.
What it has against it:
The price stays anchored. If SEO in Spain is charged between €600 and €4,000/month (pacoruben.com) and you add GEO "inside", the client will perceive the expansion as a percentage of what they already pay, not as a service with its own value. A reasonable extension moves in the +€300-900/month range; the same work sold as a standalone service to a new client supports equal or higher rates without the constant comparison with "what I already pay".
Risk of giving it away. The temptation of "I'll include it so you don't leave" is real, and once included for free, its price is fixed at zero forever.
If SEO is questioned, it all falls. A client who decides to cut SEO takes down the GEO that lived inside the same contract.
Model 2: standalone service
AI visibility as a product with its own name, price, deliverables and invoice line. It's sold to clients with SEO, without SEO, and even to people who only hired you for a website two years ago.
What it has going for it:
A genuinely new ticket. It's not a +20% on the retainer: it's a second recurring revenue source with its own price range. Ten clients at €400-500/month is €4,000-5,000 of MRR that didn't exist before — the full maths of productising it is in how to grow recurring revenue by productising GEO.
Positioning. "An agency that also does GEO" and "an agency specialising in AI visibility" don't compete in the same league. The standalone service gives you something to show: your own page, cases, a methodology. In 2026 there's still room to be "the AI ones" in your city or vertical; in 2028 there won't be.
A bigger market. You can sell it to people who'd never buy SEO from you: the client who already has an SEO agency (and doesn't want to change), the local business with no budget for €800/month of SEO but yes for €300 of AI presence, the dormant roster of old web projects.
What it has against it:
More friction. A new service means a proposal, client education and a sales cycle. You're explaining a category the average buyer doesn't know yet — though the data helps: frequent ChatGPT use in Spain has gone from 4% to 28% in two years (Funcas, III AI Survey, 2026).
It needs its own scaffolding. A delivery process, its own report, its own KPIs. You can't dispatch it with two paragraphs at the end of the SEO report; a service with its own invoice demands reporting that stands on its own.
The comparison in short
| SEO extension | Standalone service | |
|---|---|---|
| Sale | Fast (own roster) | Slow (requires educating) |
| Ticket | +€300-900/month on the retainer | €300-900+/month of its own |
| Who it reaches | Only your SEO clients | Anyone |
| Retention | Reinforces the retainer | Creates a new relationship |
| Risk | Anchored price, possible giveaway | More sales cost |
The recommendation, by type of agency
SEO agency with a roster of retainers: start as an extension — it's the fastest sale and it shields your retention — but with its own line on the invoice and an explicit price from day one, never "included". To win new clients, present it as a standalone service. It's the hybrid model, and it's the one that works best.
Web design / development studio: standalone, no question. You have no retainer to hang anything off, and this service is exactly the recurring piece your project model is missing. Your roster of delivered websites is your prospect list.
Full-service or paid media agency: standalone, packaged alongside what you already sell ("your brand, in Google Ads and in the AI answers"). The ads client gets the coverage argument quickly.
Freelancer / consultant: extension for your current clients, because your bottleneck is sales, not product. When you have 4-5 clients with the service, give it a name and its own page.
In any of the models, the tool is the same and so is the cost: with Surfeo for agencies you pay €20/month for the base account plus €35 or €79/month per client depending on plan, each client with their own space, their prompts monitored every week across the 4 AIs and their PDF report. Whether it's an extension at €300 or a standalone service at €700, the tool cost per client doesn't change — what changes is what the packaging lets you charge.
Frequently asked questions
Can I start as an extension and split it off later?
Yes, and it's the natural path: extension with an explicit price for 6-12 months, and once you have cases and a settled process, you turn it into a service with its own name for selling to new clients. The reverse (from standalone to included) is lowering the price to zero, and there's no coming back from that.
What if I include it free in the retainer to differentiate myself?
No. Give away the initial audit — the snapshot of where they appear and where they don't — as a hook, never the recurring service. What's given away once isn't charged for later, and this work has hours and a tool behind it every month.
Won't the client complain if I charge separately for something that "sounds like SEO"?
Some will ask. The honest answer works: they share foundations, but GEO measures other things (prompts and mentions, not positions), in other engines (4 AIs, not Google) and with additional work on top. What's the same and what's genuinely new we break down in GEO vs SEO: what changes.
What price do I put on each model?
As an extension, €300-600/month for an SME; as a standalone, €300-900/month depending on size and sector. The full ranges with the margin maths are in how much to charge for AI visibility services.
Whatever wrapper you choose, the first step is the same: knowing where your clients stand today. Take the free visibility test with two or three websites from your roster and you'll have the material to decide with data, not theory.
