How We Fixed Our Own Search Optimization Consultants Problem

By Trisha Seal · August 4, 2026 · Seal Global runs search, content and AI visibility programmes for service businesses, law firms, dental practices and ecommerce brands.

The embarrassing part

We sell search strategy. In the first half of 2026 we were ranking on page one for our single most valuable commercial term — search optimization consultants, roughly 9,900 monthly searches, keyword difficulty 38, a $56 cost per click — and it was producing almost nothing.

Not zero traffic. Traffic was fine. It was producing almost no enquiries. For a term where a single click costs fifty-six dollars in Google Ads, that is not a rounding error. That is money on the floor, in our own house, in the exact discipline we charge for.

What follows is the whole thing: what we found, what we changed, what happened, and the part that did not work. We are publishing it because it is the most common failure we find in client audits, and because the fix costs almost nothing compared with the volume of new content most firms throw at the problem instead.

The diagnosis: right keyword, wrong URL

The term was ranking. It was ranking on a blog post. A long, genuinely useful blog post explaining what search optimization consultants do and how to evaluate one.

That sounds like a win. It is not, and the reason is intent. Someone typing that phrase into Google at a $56 CPC is not curious. They are shopping. They want to know what it costs, what they get, who they would be working with, and how to start a conversation. A blog post hands them a reading experience and a byline. They read it, they nod, and they go buy from somebody with a services page.

Three secondary problems compounded it. Our internal links pointing at that topic were split across four different URLs, so no single page accumulated authority. The blog post carried Article schema, which tells a machine "this is an essay", not "this is a purchasable service". And because the site renders in the browser, an AI crawler that does not execute JavaScript was seeing a near-empty document where the structured data should have been — the exact failure mode that answer engine optimization is meant to prevent.

The fix, in the order we did it

1. Built a top-level service page at /search-optimization-consultants — commercial intent, transparent pricing bands, named deliverables, scope, and an obvious way to start. Not a landing page. A real page with the depth of the blog post plus everything a buyer needs.
2. 301-redirected the blog URL into it. Not a canonical, not a soft link — a permanent redirect, so every signal the old URL had earned transferred rather than competing.
3. Consolidated internal links. Every mention of the phrase across the site now points at one destination with consistent anchor text. This is the least glamorous step and one of the highest-leverage.
4. Moved the schema into server-rendered HTML. Service, Organization, FAQPage and BreadcrumbList JSON-LD are now in the static document at build time, so a crawler that never runs JavaScript still gets the full picture.
5. Rewrote the first 120 words to answer the query directly, in plain language, before any narrative. That is what an AI answer engine lifts.
6. Started tracking citation share across ChatGPT, Perplexity, Gemini and Google AI Overviews on a fixed prompt set — separately from Google position, because they move independently.

What actually moved

The redirect was recrawled within days. Position on the consolidated URL was stable inside three weeks — we did not lose the ranking, which is the fear that stops most people from doing this. The meaningful change was downstream: the same visibility started producing conversations, because the page finally matched what the searcher came to do.

The AI side moved on a different clock. Citation share improved once the structured data was server-rendered and the entity information was consistent, which took roughly six to ten weeks — much closer to how generative engine optimization behaves generally than to how classic rankings behave.

The part that did not work

Our first instinct was wrong. Before the diagnosis we did what most teams do: we published more content around the topic. Three additional posts. It made things measurably worse, because four pages chasing overlapping intent is cannibalisation, and Google resolved the ambiguity by trusting all of them slightly less.

Publishing more is the default reflex because it feels productive. It is also the most reliable way to dilute a term you were already close to owning. If you take one thing from this: before you write anything new, check whether the pages you already have are fighting each other.

What this means for your site

This pattern — high-value commercial keyword landing on an informational URL — is the single most common finding in the audits we run, and it shows up just as often inlocal SEO setups as in national ones. It is also the cheapest to fix, because you are not creating demand, you are stopping yourself from leaking it.

Run the check yourself. Pull your top twenty organic terms by commercial value, not by volume. For each one, look at which URL ranks and ask whether that page lets someone buy. If the answer is no on more than a couple, you have the same problem we had. Good search optimization consultants spend more time on that question than on keyword discovery, because the gap between ranking and revenue is almost always structural rather than a content shortage. The same audit logic sits underneath ourgenerative engine optimization services and our broader AI search optimization work, because a page that cannot be bought from is a page an answer engine has little reason to cite.

Related reading

• Search optimization consultants: the complete 2026 hiring guide
• 8 search optimization consultant myths that cost businesses rankings
• Generative engine optimization: the complete guide
• The local SEO services playbook