A new section for development buyers on the site of a global IT services company, one search engines hadn’t found yet, in a niche where you compete with the established domains of global outsourcers. Over 20 months the section grew from zero to about 10,000 search visits a month, by Ahrefs’ estimate.
Results at a glance
Search visits per month
0 → ~10,000
Ahrefs estimate for the new section of the client’s site over the period of our work. At the start the section had neither traffic nor queries in the Top 10.
Queries in Google’s Top 10
0 → 1,500+
More than 300 of them in the Top 3.
Traffic and rankings come from Ahrefs’ public data for the period of our work. Under the terms of our contract we don’t name the client or the section’s address, which is why the figures are rounded.
Before
A new section search engines hadn’t found yet
| Client | a global IT services company, a new section of its site for development buyers |
| Niche | software development and dedicated teams for businesses |
| Markets | US and English-speaking markets |
| Services | SEO, content briefs, link building, analytics |
| Period | 20 months, 2021–2023 |
| Status | completed |
| Team | 6 specialists and a project manager |
The client’s remote teams platform had two lines of business, and it’s important to tell them apart. One hires engineers (people apply, go through interviews and tests, join the system and then get put forward for projects). The other, the new section of the site, sells the teams themselves to companies that need software built. We worked specifically with the B2B branch.
The section was essentially new, so for the first months we worked with the existing structure and put the rebuild off.
The goal
Enquiries from companies that need a team
A fintech startup in the US wants to launch its own payment system. It doesn’t have a development team yet. What it has is an idea, investor money and the need to build the product reliably, fast and well. Those were the people who had to find the client on Google and send an enquiry.
So the goal was enquiries from companies that need development work. The difficulty was that in SEO terms the section was brand new, and it had to compete with the established domains of global outsourcers.
What we did
Four verticals first, then articles and hiring pages, then commercial pages
Months 1–5
Four verticals out of the whole section
We focused on four verticals out of the whole section and laid out a structure around buyer intent.
Four verticals out of everything
Effort spread across every possible direction wouldn’t have moved any of them. So we took four verticals (Retail, Big Data, Fintech, Healthcare) and regrouped queries and pages around them.
A structure built on buyer intent
Someone searching for “PHP development company” wants a project built for them, while someone searching for “hire PHP developers” wants people for their own team, and search shows different pages for those queries. We left the existing structure as it was for the first months so the results wouldn’t be delayed, and gave recommendations for later (child pages for services, industries and solutions, a separate hiring branch for “hire … developers”, an internal linking scheme).
Services and “how to” articles
A founder who needs a payment system first searches for how to build one and what it costs, and only then for “who will build it”. So we built the service pages on standard, well-tested templates (vertical, service, outstaffing, technology), including pages for startups and the first hiring pages, and the articles were written around buyers’ “how to …” queries (banking and fintech apps, MVPs and the like). The content plan was split down the middle; top-of-funnel topics fed the existing pages, and bottom-of-funnel topics were prepared for new ones.
Brand rules in each brief
We checked each brief against the client’s brand rules and checked each article twice, for SEO and for marketing. We also looked at whether the text spoke to the person ordering development (a startup founder, a small business, a corporation).
Betting on guest posts
For a new section in a niche full of global outsourcers, authority was the bottleneck, so we bet on guest posts. We filtered the sites for them against our checklist, dropping sites with a bad neighbourhood and keeping the ones with good, smooth traffic dynamics, with no sharp jumps in their link profile and none of the other flaws that make a site low-quality.
What the client did. Implemented our briefs, service pages first, then articles. Later we switched to writing all briefs in English.
Outcome. The section showed up in search, and the Ahrefs estimate went from zero to about a hundred visits a month. There were no queries in the Top 10 yet, so the stage’s result was a framework of pages and the first articles in the index.
Months 6–9
A plateau before the release of the new blog and a menu built on our structure
Traffic was growing, but more slowly than it needed to. Rankings of the important pages had barely moved, so the section could only grow on new pages and better rankings for the pages it already had. That’s why we kept handing over briefs, and when the new blog and menu went live, the backlog of pages went onto the site.
Briefs in advance
Sometimes weeks pass between a brief and a page in search, and if new briefs wait until the previous ones are out, the section sits for another month or two with nothing new. So we handed over briefs in monthly batches (healthcare, fintech, retail, travel, banking and telecom). When the client rolled out the new blog, the backlog of pages went live with it.
A new menu built on our structure
The menu and the nesting of sections tell search which pages are main and which are children, and pass weight to them through internal links. When the client launched a new menu built on our recommendations, we checked it together with the client’s developers. Alongside that we started briefs for internal linking, so links between pages would complement the menu.
What the client did. At the end of the stage it rolled out the new blog, the new menu and the section nesting.
Outcome. The Ahrefs estimate held at a few hundred visits a month. Rankings rose after the links, though without reaching the Top 3 or bringing noticeable traffic yet; that was the baseline.
Months 10–13
An almost fourfold jump when the backlog of briefs and the new blog went live, and the discovery that the site was competing with itself
Together with the new blog, the backlog of service pages produced a jump. After every release we checked the site together with the client’s developers so the new pages would get indexed. Along the way it turned out that job listings for engineers were in places competing with the hiring pages, because to search the phrase “Hire … engineers” sounded almost the same on a listing and on a hiring page. We started separating them.
Calls to action for each reader’s stage
Readers of the same article are at different stages; one isn’t ready to talk to sales but will happily work out a cost, while another is ready to describe the project. So we proposed several types of call to action, each for its own stage.
The site competing with itself
To search, a “remote … developer jobs” listing and a “Hire … developers” page sounded almost the same, and they lived in the same folder of the site, so for buyer queries search sometimes showed the job listings (and that is cannibalisation). Using examples, we showed that “hire …”, “freelance …” and “remote … jobs” queries were mixed together in the results. Towards the end of the stage we ran a test on one profession, de-optimising the listing and re-optimising the hiring page.
Clusters around the hiring pages
A cluster of articles around one commercial page answers the related questions, links to the page and builds Topical Authority, but it works once there are enough articles in it. So when the first hiring pages appeared, we built clusters around them one at a time (outsourcing, offshore and dedicated teams, comparisons with neighbouring languages and frameworks, development costs).
What the client did. Published the backlog of briefs and updated the calls to action in articles already published.
Outcome. The Ahrefs estimate grew from a few hundred to over a thousand visits a month, almost fourfold, and the jump coincided with the backlog of briefs going live together with the client’s new blog and menu.
Months 14–16
Hiring pages, article clusters and separation from job listings brought almost three times the traffic
In short, we separated job listings from hiring pages at scale and pointed links mainly at the hiring pages. Traffic grew almost threefold.
Separating the pairs at scale
Our test on one profession showed how to separate a job listing from a hiring page, and the site had as many such pairs as it had professions. So we separated a few professions a month (de-optimising the listings, re-optimising the hiring pages and adding the new pages to the internal linking).
Links mainly to hiring pages
Hiring pages are commercial, and getting one into the top results is harder than with an article, so links matter more for it. That’s why we pointed links mainly at them.
What the client did. Kept launching new hiring pages.
Outcome. The Ahrefs estimate grew from over a thousand to about 3,000 visits a month, almost threefold; the data shows no single cause for the jump, which coincided with the hiring page launches, the article clusters and the links to hiring pages.
Months 17–20
Traffic reached its peak
In the last months we pushed the hiring and service pages over the line where it was needed, and protected what had been earned.
Service pages
Commercial pages bring in people who are already looking for a contractor, so we wrote briefs for new service pages and for re-optimising existing ones with new blocks (services and industries, skills, technologies).
What the client did. Implemented the briefs for the hiring pages and the service pages.
Outcome. The Ahrefs estimate grew from about 3,000 to about 10,000 visits a month. The section had over 15,000 ranking queries and over 1,500 in the Top 10, more than 300 of those in the Top 3 (zero at the start). This was the cumulative effect of the articles, hiring pages and links from our joint work.
What was hard
A new section among established domains, and a big site that kept changing
The section had no authority yet
In the global outsourcing niche the top spots have long been held by established domains, and the new section had no authority yet. So we didn’t spread ourselves across the whole section; we took four verticals and built links through guest posts on vetted sites.
The site was competing with itself
To search, job listings for engineers and pages for buyers sounded almost the same, so for buyer queries it sometimes showed the listings. We first tested the separation on one profession, and then separated the pairs a few professions a month.
A big site kept changing
On a big site with frequent releases, every release is worth checking. So after every release we checked the site together with the client’s developers, kept technical fixes in a separate priority table and checked every month which recommendations had appeared on the site.
After
A section that effectively didn’t exist became a channel for enquiries
| Before | After |
|---|---|
| Ahrefs saw no traffic or queries in the section | About 10,000 search visits a month and over 15,000 ranking queries |
| No queries in the Top 10 | Over 1,500 queries in the Top 10, more than 300 of them in the Top 3 |
| A section that was essentially new to search | Four verticals, fintech articles built around buyers’ queries, separate pages for engineers and for buyers |
Why it worked
Three decisions that shaped the result
A narrow front
We took four verticals out of the whole section and went deep on them, and within five months the section showed up in search.
Topics where buyers start
Articles on how to build a payment system brought in people who were already working out a budget. A founder first searches for how to build the product and only later for a contractor.
Separated intents
We separated job listings for engineers from pages for buyers by intent, and for buyer queries search started showing the hiring pages.
Takeaways
Four lessons for anyone selling development work
- On a new section, narrow the front. Effort spread across every direction won’t move any of them. Here we took four verticals out of the whole section, and within five months the section showed up in search.
- Write about what the buyer wants to build. A founder who needs a payment system searches for “how to create a payment system” first and looks for a contractor later. That article brings in someone who’s already doing the maths.
- Check whether your site is competing with itself. If one domain has both “we’re hiring” and “hire from us” pages, separate them by intent before search picks the wrong one.
- On a big project, check everything after each release. A release on a big site touches many pages at once, which is why it’s essential to check with specialised tools (Screaming Frog, Netpeak Spider, Google Search Console) that things are going the way they should. And to react quickly when something is off.
Questions
What people ask about this case
How long did it take to see results?
The first months went into the foundation (four verticals, the first articles, the first links), and within five months the section showed up in search with its first hundred visits a month, by Ahrefs’ estimate. The biggest growth came in the second year, when the backlog of briefs, the hiring pages and the article clusters around them went live.
Can these numbers be checked?
Traffic and rankings come from Ahrefs’ public data for the period of our work. Under the terms of our contract we don’t name the client or the section’s address, which is why the figures in this case are rounded.
Why de-optimise your own pages?
When one site has both job listings for engineers and an offer to hire engineers, search can show the wrong page to the wrong people. So we separate them by intent; we de-optimise the listings for buyer queries and re-optimise the hiring pages, a few professions a month. Other clashes of this kind come up from time to time as well.
