FAQ
Seventeen straight answers, including the awkward ones
Every question here is one a real buyer has a right to ask before spending money with a new agency. Where the honest answer is bad for us, we have left it in.
Guarantees and results
The short version: we do not guarantee anything about placement in AI answers, because no agency on earth can. What we guarantee is the work, the measurement method and the fact that you own everything we build.
Can you guarantee my business will appear in ChatGPT?
No. Nobody can, and anyone who says otherwise is either guessing or lying. AI engines choose sources at generation time using signals no agency controls. Google and Bing both tell buyers to walk away from anyone guaranteeing rankings. What we can do is remove every technical reason you are ineligible, then measure honestly.
What results can you actually promise, then?
Three things. That your site becomes technically eligible to be cited. That you get honest measurement across named consumer surfaces with repeated observations. That every page, schema block and profile we create belongs to you permanently. Outcomes in the answer boxes themselves are influenced, never promised.
Is AI search actually worth paying for yet?
AI referral traffic is roughly 1% of most sites' visits today, so no, not as a traffic channel. The commercial damage is on the other side. A randomised field experiment in 2026 found organic clicks fell 39.8% when an AI Overview appeared. You are losing traffic you already paid for.
What is a realistic timeline before anything changes?
Technical eligibility fixes land within days. Entity and structured data signals typically settle over four to eight weeks. Answer coverage and third-party citations are the slow part: expect three to six months before appearance frequency moves meaningfully. Anyone promising faster is selling you a story.
Method and measurement
How the work is done and how you can tell whether it worked. The full method page covers all of this in more depth, including the research and the techniques we refuse to use.
How is this different from SEO?
About 80% of it is conventional SEO done properly: crawlability, structured data, real content, local signals, reviews. The other 20% is genuinely new: per-agent crawler rules, server-side rendering for engines that cannot run JavaScript, answer-shaped pages, and repeated prompt measurement across named consumer surfaces.
Why won't you give me a single visibility score?
Because independent measurement shows only about a third to two-fifths of cited sources stay the same between identical runs. A single number hides that variance behind false precision. We run every prompt at least three times and report appearance frequency with a confidence interval instead.
Which engines do you actually track?
Five: ChatGPT, Google AI Overviews and AI Mode, Perplexity, Claude and Microsoft Copilot. The exact consumer surfaces used are stated in the order and report. If a sixth becomes commercially relevant we will add it and say so in your monthly report.
Does llms.txt do anything?
No, as far as anyone can currently demonstrate. Google has publicly stated Search does not use it and no AI platform documents consuming it. We add one anyway because it takes twenty minutes and harms nothing, but it is not a line item and we will not charge you for it.
Do you use any techniques that could get me penalised?
No. No hidden text, no prompt-injection tricks aimed at crawlers, no bought links, no AI-generated content published at volume to game retrieval. Those carry real manual-action risk and a short half-life. Everything we do is defensible if a search engineer reads your page source.
Money and contracts
All prices are published on the pricing page and exclude VAT. Audits are one-off. Retainers are rolling monthly with no minimum term, no notice period and no exit fee.
Why do you publish your prices?
Because the category is full of opaque retainers that start with a discovery call designed to price you rather than help you. Published prices let you decide in ninety seconds whether we are worth a conversation. It costs us some leads and saves you and us a great deal of time.
Do I have to sign a long contract?
No. Both retainers are rolling monthly and you can cancel any month with no exit fee. Everything we build — pages, schema, profiles, documentation — belongs to you and stays on your own domain and accounts. There is nothing to hold hostage if you leave.
Is the Deep Report really credited in full?
Yes. Buy the £595 Deep Report, start Core or Growth within 30 days of delivery, and the first Core invoice is £0, or the first Growth invoice is £855. The report is credited in full. If you do not start a retainer you keep the report, the raw run data and the ninety-day roadmap regardless.
What if my site is already fine?
Then we will tell you and you will have spent £195 to find out, or nothing at all if the free scan settles it. That happens occasionally. It is a much better outcome than a twelve-month retainer that quietly bills you for work your site did not need.
Working together
We are a small, remote, founder-led UK agency. That shapes what we are good at and what we cannot do, so it is worth being clear about both before you commit.
I already have an SEO agency. Does this conflict?
Usually not. We work alongside them and send them the technical fix list if you prefer. Our work sits on top of good SEO rather than replacing it. If your current agency is already doing schema, rendering and review work properly, we will tell you and scope down accordingly.
Why should I trust an agency with no case studies?
You should not trust us on reputation, because we have not earned one yet. Trust the checkable things instead: a published method, published prices, a £195 entry point, no lock-in, and our own scores published monthly. That is a harder standard than a testimonial you cannot verify.
You work remotely. Does that matter?
Only if you need someone physically on site. Scoping, audits, fixes, reporting and calls are all done remotely, in UK hours and in pounds. If you want a person walking into your practice each month, hire locally and we will tell you so rather than take the work.
What data do you collect from the free scan?
The URL you enter, plus the optional business name and town. Nothing else, and no email unless you choose to request the full fix list. We do not sell or share submissions, and you can ask us to delete yours at any time. Details are in the privacy policy.
Question not answered here? Email hello@citemark.co.uk and you will get a written answer from the person who would do the work. If it is a good question we will add it to this page.
Start here
The fastest way to answer the real question
Which is: can AI engines see your site at all? Sixty seconds, free, score shown before we ask for anything.