Frequently asked questions

GEO is the work of making your brand a cited source inside the answers generated by engines like ChatGPT, Gemini, Perplexity and Google AI Overviews.

Classic SEO aims for a position in a list; GEO aims to be inside the single answer. The two share the same foundation, but GEO has its own requirements: entity clarity, quotable paragraph structure, machine-readable access, and measurement of answer share.

We build a fixed list of target questions, run them across each engine at regular intervals, and record whether your brand is cited as a source.

The report shows three things: which questions you appear in, which competitors take your place, and how that changes over time. We repeat the measurement monthly — a one-off check means little, because results shift as the models are updated.

llms.txt is a plain-text file that tells large language models what your site is about and which pages matter most.

It is not a required standard like robots.txt — it is an emerging convention. The cost is low enough to be worth adding, but it will not deliver visibility on its own. What actually decides the outcome is whether your pages serve clean server-rendered HTML and whether your robots policy leaves AI crawlers any access at all.

On informational queries you will likely lose some clicks; but when you are the source of the answer, the traffic that remains converts better.

The queries hit hardest are short definitions and how-tos. On comparison, pricing and purchase-intent queries people still come through to the site. The right response is to separate those two groups: aim to be the cited source on informational queries, and to strengthen the page on commercial ones.

If you want to be visible, no — you cannot appear in the answer of an engine you have blocked.

The decision splits in two: having your content used as training data and being cited when an answer is generated are different things, and they can be managed separately in robots.txt. For most brands the right approach is to stay open to search-oriented crawlers while restricting training-oriented ones by preference. We set that distinction up for you.

Pricing follows site size, competitive intensity, number of target markets and monthly production volume. The plans are a starting point; after the first call we scope the work and quote a fixed monthly fee.

Yes. Deciding not to work together is a perfectly normal outcome. The report is yours to use either way. The point is that both sides see the size of the job before committing.

In most projects between six and nine months; sooner on sites with little technical debt and existing authority.

The calculation is straightforward: search volume on target terms, a realistic click-through rate, your current conversion rate and average customer value. With those four we work out the expected return together on the first call. If the numbers do not hold up, we say so upfront — not everyone needs SEO.

Content production is included in the plan; on links there is no buying, so there is no separate link budget either.

Every plan includes a set number of content briefs or fully produced pieces each month, varying by tier. Links are earned through digital PR, data studies and content with genuine editorial value. We have seen bought links work briefly and then become a liability at the next update, so we do not use them at all.

We start with a 30-minute intro call, then prepare the free audit and walk through the findings together. If we proceed, weeks one and two are full diagnosis, weeks three and four the roadmap.

We deliver findings as implementable tickets: what to change, why, and how to test it. We can join your sprint ceremonies or simply hand over the written work — your call.

We do not sell hours; we work against scope and deliverables.

Every project has one named specialist, with colleagues from the technical, content and data side joining as needed. You see what was done through the work list agreed at the start of the month and the outputs delivered at the end of it — not through a timesheet. There is no account manager in between; you talk to the person doing the work.

Every change that is visible on your site or structural goes through your approval first.

Content publishing, title and meta changes, redirects and template work all move through an approved list. Work on our own side — measurement setup, report structure — does not wait for sign-off. If you would rather not grant access, we hand every change to your own developers as a ready-to-apply list.

Technical SEO is the infrastructure work that lets search engines crawl your site completely and index it correctly.

It covers four areas: crawl budget and robots handling, indexation and canonical decisions, page speed through Core Web Vitals, and structured data plus information architecture. If this layer is weak, content and link work cannot pay off — how well a page is written does not matter if it is never crawled.

Yes, but not on their own; they act as a tie-breaker between two otherwise comparable pages.

The real effect is indirect: on a slow page people bounce, conversions drop, and the signals that page earns get weaker. We measure LCP, CLS and INP in field data (CrUX) rather than lab tests, because that is the data Google actually uses.

A migration lives or dies on its redirect map: every old URL must reach its new counterpart in a single 301 hop.

Before the move we take a full inventory of existing URLs with their traffic and link value. On launch day we update redirects, canonicals, the sitemap and internal links together, then watch crawl logs for the next four weeks. The most common cause of loss is redirect chains and forgotten deep pages.

E-E-A-T stands for experience, expertise, authoritativeness and trustworthiness — Google's quality evaluation framework.

On a site it looks like this: author pages that show who actually wrote the content, verifiable sources and citations, visible update dates, clear contact and company details, and third-party mentions. For topics like health and finance these signals are not a bonus — they are the price of entry.

No — Google looks at whether content carries original value, not at how it was produced.

Problems come from unchecked production: unverified claims, restatements of what everyone else wrote, text with no first-hand experience in it. We use AI for drafting and research, then have someone who knows the subject verify it and add field data. Content gets cited because it has something new to say, not because of how it was written.