A case study in building SEO from scratch at a B2B eMobility company with no keyword strategy, no targeting, and no organic roadmap. The content-first method that moved priority keywords up an average of 12 positions and drove an 18% lift in organic traffic from the blog, plus the honest limits of what this approach can and can’t fix.
Building SEO from scratch is a different job from improving SEO that already exists, and the moves with the biggest payoff aren’t the ones a technical audit flags first. When I joined a B2B eMobility company as its Product Marketing Manager, it had never done SEO: not badly or half-heartedly, but not at all. There was no keyword strategy, nothing on any page targeting search intent, no organic roadmap, and nobody who had ever owned search. The website existed, content went out when someone felt like it, and whether any of it ranked was left to chance.
That’s a more common starting point than the SEO industry likes to admit. Most guidance assumes a company with a foundation that needs optimising: tighten the technical set-up, chase incremental gains, fix what’s underperforming. At a company starting from zero you’re creating a system where none existed, across, in my case, 8 eMobility segments and 39 countries.
What does a company with no SEO look like?
Like a site that works fine for everything except search, which is why nobody notices: no page targeted a keyword on purpose, and titles read as creative headlines with no relationship to what anyone searched for, and meta descriptions were either generated from the first line of body text or missing. There was no keyword research anywhere, so nobody knew what the company’s buyers typed into Google, and the team wrote content around whatever it found interesting.
Pages competed with each other by accident, two or three half-targeting the same vague idea and none of them ranking. The blog was a publishing habit with no strategy behind it.
None of this looks broken from the inside, because the site loaded fine, content went out, and traffic existed, mostly direct and referral, so nobody felt the absence of organic search: you don’t miss visits you never knew you could have had.
Why start building SEO from scratch with content instead of a technical audit?
Because this site’s problem was relevance, and a technically flawless page that targets nothing ranks for nothing. The instinct, when you’re handed a site with no SEO, is to run a technical crawl first: broken links, missing alt text, page-speed scores, crawl errors. But the pages loaded fine and Google could crawl them; what Google couldn’t do was work out what any page was about, because no page had been written to be about anything searchable. Google’s own SEO Starter Guide makes the same point from the other side: for most sites, “You usually don’t need to do anything except publish your site on the web” to be found.
So I went content-first, putting structure and intent ahead of technical housekeeping. A technically average page written to match a query with buyer intent can rank, and in a niche where the competition is thin it can rank quickly; with limited time and one person doing the work, I put the effort where the ceiling was highest. That isn’t the order every site needs:
| The site’s binding constraint | Start with | How you can tell |
|---|---|---|
| Relevance: no page is about anything searchable | Content: keyword map, titles, headings, intent | Pages index fine but rank for nothing useful |
| Technical: Google can’t crawl or index the pages properly | A technical audit | Pages missing from the index, crawl errors, very slow loads |
| Authority: good pages lose to bigger domains | Links and digital PR | You rank on page two for terms dominated by large sites |
The sequencing flips for AI-era discoverability, where structure carries more weight, and I dug into that contrast in my AEO and LLMO framework and in how to sequence SEO, AEO and LLMO.
How do you build a keyword strategy for a niche B2B industry?
Segment by segment, and aimed at low-volume, high-intent terms instead of head terms. eMobility is a specialised B2B space with no million-search keywords to chase, and the few high-volume ones that exist are generic, low-intent and dominated by much bigger domains. The opportunity lives in the specific terms a charge point operator, a fleet manager or an eMSP types when they’re evaluating a solution. That long tail is most of search anyway: in Backlinko’s study of Google click-through rates, 90.3% of queries had 10 or fewer impressions.
The company sold into eight segments, from energy providers and charger manufacturers to charge point operators, fleets and eMSPs, and each had its own vocabulary and its own buying questions. For each one I mapped the terms that segment used, separated buyer-intent queries from noise, and identified where the company had a fair chance to rank given its modest domain authority. The result was a keyword map that gave every segment a clear set of targets, which the company had never had. Running that across eight segments as one person leaned on the same discipline I wrote about in multi-persona content strategy as a solo marketer.
Which on-page changes moved rankings when building SEO from scratch?
Unglamorous ones: titles, meta descriptions, headings and content rewritten to match the intent behind a target query. The work split between fixing existing pages and building new ones.
On existing pages, I rewrote title tags to lead with the term the page should rank for, instead of a clever headline that targeted nothing, and wrote meta descriptions to earn a click instead of describing the page to nobody. Google’s starter guide is plain about both: think about the words a user might search for, and keep each description short and unique to its page. I restructured headings so the H1 and H2s signalled the page’s topic, and reorganised body content to match the intent behind the query instead of wandering through whatever the original writer wanted to say.
On new content, the blog did most of the work, and I built each piece around a mapped keyword and a question from one of the eight segments, structured so the answer was clear. I produced 12+ articles, 15+ web pages, two whitepapers and a case study, each pointed at a specific search intent, and internal linking tied the segments together so the site read as a connected resource instead of a pile of unrelated posts. The same audience research fed the launch work, which I covered in product launch content on zero budget, and the social rebuild I ran in parallel, in the social turnaround that lifted reach 217%.
Rewriting existing pages tends to move faster than publishing new ones, which helps on a short timeline: Ahrefs’ 2025 study of ranking pages found only 6.11% of new English pages reached Google’s top 10 within a year.
What does a 12-position average climb mean in traffic?
It means a share of priority keywords crossed from page two to page one, which is where organic clicks begin. An average climb of 12 positions sounds incremental until you see how top-heavy click-through is: Backlinko’s analysis of about 4 million search results found the number one organic result earns about 27.6% of clicks and is roughly 10 times more likely to be clicked than the result at position ten, while only 0.63% of searchers clicked anything on page two.
A keyword moving from position 20 to position 8 goes from earning almost nothing to earning clicks. Averaged across a set of priority keywords, a 12-position climb meant a good share of them made that jump, and that’s the kind of movement behind the 18% growth in organic traffic.
What would I do differently with more time?
Mostly, I’d measure better and widen the scope sooner.
I’d build a proper measurement baseline on day one: I tracked the climb and the traffic lift, but cleaner before-and-after data per segment would have made the wins easier to prove and the strategy easier to argue for internally, which is why I’m so particular about the content marketing metrics that get reported.
Once the content foundation was in, I’d push harder and sooner on the technical and off-page side. Content-first was the right opening, but SEO compounds, and building authority and tidying the technical layer is where early wins either accelerate or stall. I’d also spend more time on competitor keyword gaps, because a deeper look at where competitors were winning and we weren’t would have sharpened the target list.
The obvious limit on building SEO from scratch there was time: I was there for eight months before the company went bankrupt, which I wrote about in a short tenure on a CV, so I saw the climb and not what compounding would have done in year two.
What can’t content-first SEO fix?
Serious technical problems, domain authority, and a small market’s traffic ceiling. Content-first is the right call when a site’s core problem is relevance, which is the usual situation at a company that’s never done SEO, but it isn’t universal. If a site has serious architecture, indexation or speed problems, well-targeted content won’t rank until you fix the technical layer. It can’t manufacture domain authority either: on competitive head terms, the page that wins is often the one on the domain with the backlink profile to match. And in a small niche, content-first SEO gets you to the top of terms that don’t have much volume, so the absolute traffic ceiling stays modest however well you rank.
Search itself has also changed since this work: in SparkToro’s analysis of US Google searches in early 2026, 68.01% ended without a click, and Pew Research found users clicked a result in 8% of visits when Google showed an AI summary, against 15% without one. Rand Fishkin’s advice is blunt: “Our belief and advice is to invest in Zero Click Marketing”. Page one still beats page two by a wide margin, but the same ranking earns fewer clicks than it did in 2024, which is why I’d now pair the keyword map with tracking impressions that don’t turn into clicks and a plan for organic traffic decline.
Frequently asked questions
Start with relevance. Research what each buyer segment searches for, prioritising low-volume, high-intent terms over generic head terms, and build a keyword map that gives every segment clear targets. Then rewrite existing pages so titles, meta descriptions and headings match those terms, and build new content around the map. At a B2B eMobility company that had never done SEO, this content-first approach moved priority keywords up an average of 12 positions and grew organic traffic 18%.
It depends on the site’s binding constraint. If pages index fine but rank for nothing useful, the problem is relevance, so start with content. If Google can’t crawl or index the pages properly, start with a technical audit. If good pages keep losing to bigger domains, the gap is authority, which needs links. Google’s own starter guide notes that most sites don’t need special technical work to be found, so relevance is the usual first fix at a company starting from zero.
Research segment by segment, because each buyer segment has its own vocabulary and buying questions. Skip the few high-volume head terms, which tend to be generic and dominated by large domains, and target the specific terms a buyer types when evaluating a solution. Low-volume keywords are most of search: in Backlinko’s data, 90.3% of queries had 10 or fewer impressions. Prioritise the terms where your domain has a fair chance to rank.
Rewriting existing pages for the right terms can move rankings within weeks to months, while brand-new pages take longer: Ahrefs found only 6.11% of new English pages reached Google’s top 10 within a year. Google’s own guidance says some changes take effect in hours and others take several months. Expect the first gains from fixing what’s already published, and plan for authority-building to take longer.
