How Generect went from zero rankings to 3.7× search clicks in three months

Starting a blog from zero rankings

Generect is a B2B lead search engine. Sales teams, founders and agencies use it to find verified contacts, check emails and build prospect lists. It also has an API and an MCP server, so AI agents can use it directly.

The product was solid, and the team had momentum. In 2025, Generect was one of 35 startups picked for the Google for Startups Ukraine Support Fund.

What they didn't have was search traffic. We ran a rank tracker on their core keywords from December 2024 to March 2025. Every single one sat at position 101 – not in the top 100 at all. That included terms like "b2b leads" (1,300 searches a month) and "b2b database" (500).

There was no blog yet. Just a product site, competing against some of the biggest names in B2B data:

• ZoomInfo, Apollo, Lusha and Clay all run large content operations with years of backlinks behind them.
• The head terms are brutal. "B2B data providers" is a fight a new site loses for a long time.
• The buyers are sceptical. Sales people have tried many data tools. Generic "top 10" posts don't move them.

The usual paths didn't fit a small team:

We needed topics where a new site could win quickly, and content good enough that the people who found it would try the product.

What we built

This was more than articles. It was the whole system that makes a blog work.

• The blog itself. In March 2025 I wrote the spec: WordPress on generect.com/blog rather than a subdomain, so every article builds the main site's authority, plus a sidebar with the author and table of contents, and a built-in CTA system.
• The content plan. A sheet with one row per article: the keyword, the stage in the buying journey, what we wanted the reader to do next, and an outline. Every outline ends with a section on how to do the job with Generect.
• 46 published articles, April to November 2025.
• Landing pages. An email finder page and six "lead generation for {industry}" pages: software, financial services, manufacturing, IT and MSPs, consulting, recruitment.
• Lead magnets. A prospect list building checklist, with its own landing page and emails.
• Links and listings. 36 editorial backlinks and 18 live directory listings.

The bet that worked: MCP content

In April 2025, most B2B data companies weren't writing about MCP (Model Context Protocol). It was a developer topic. Search volume was small and the keyword tools said the traffic potential was low.

We wrote about it anyway, because it fit Generect exactly. Generect was launching its own MCP server. The people searching for MCP were building AI agents, and those agents need contact data.

So the first cluster was:

• What is MCP?
• The n8n MCP guide
• A list of 45+ MCP apps
• Claude MCP servers, HubSpot MCP server, Salesforce MCP
• Firecrawl MCP, LangGraph MCP client
• Later: Zapier MCP, Pipedream MCP, OpenAI Agent Builder with Generect MCP

Because nobody big was covering these topics yet, a new site could rank for them quickly. MCP content became the biggest source of non-brand clicks on the blog. Search terms like "langgraph mcp client", "n8n mcp tutorial" and "hubspot mcp server" brought steady clicks while the harder terms were still climbing.

The clusters after that

Once MCP content had earned the site some trust, we moved closer to what Generect sells:

Most articles ran 3,000 to 5,000 words. That length was deliberate. API reviews and comparisons have to answer technical questions properly, or a developer or sales leader closes the tab.

How the articles were made

We planned, briefed and edited every article. Some drafts were written by outside writers working from our briefs and rules. Two rules mattered most:

• Every outline ends with the product. The last section shows how to do the job with Generect. It's placed where the reader has just learned the problem, not squeezed into the introduction.
• No lazy AI writing. Drafts were checked, and anything that read as generated was rewritten.

The planning method became a reusable tool. We later turned it into an n8n workflow that builds content plans the same way.

Getting the blog noticed

New sites rank slowly without links. So we earned them:

• 36 editorial backlinks between September and December 2025, from sites like termly.io (DR 92), Podcastle, Reply, ColdIQ, EngageBay, Salesmate and Coupler.io. Several came from companies in my own client network, which had real reasons to cite Generect's articles.
• 18 live directory listings, including G2, Trustpilot, SaaSHub, AlternativeTo and SourceForge. We applied to about 30 and skipped 32 more that charge for a listing.

What I delivered

• Homepage copy rewrites (2024)
• The blog architecture spec
• A content plan with 90+ rows: 46 published, the rest queued
• 46 articles, briefed and edited
• An email finder landing page and six industry landing pages
• A lead magnet with its landing page and emails
• 36 backlinks and 18 directory listings
• The Google for Startups announcement post

3.7× search clicks in three months

Generect started with no rankings at all. Every tracked keyword sat outside the top 100 from December 2024 to March 2025.

In three months, search clicks went up 3.7×. These are Generect's own Search Console numbers, compared over two windows of the same length:

The blog did almost all of that. In the later window, the top pages were the MCP guides: LangGraph MCP (316 clicks), n8n MCP (248) and Claude MCP (188).

Growth kept going after that. Clicks peaked at 2,757 in February 2026, the best calendar month of the engagement.

By September 2025, Generect ranked in the top five for commercial terms that had been out of reach six months earlier:

• "b2b data providers" – position 4
• "lead generation companies for small businesses" – position 3
• "firmographic data providers" – position 3
• "b2b search engines" and "lead search engine" – positions 3 to 4

What worked, and what comes next

What worked: starting where a new site could win. MCP content ranked fast because the big players weren't there yet. That early traffic helped the rest of the blog grow.

What we'd plan for: a blog in a fast-moving topic like MCP needs steady updates. New articles on these topics appear every month, so the guides that ranked first need regular refreshes to stay there.

What takes longer: the biggest commercial terms. "Lead generation database" gets thousands of impressions but sits around position 26. "B2B data provider" sits around 39. Those take longer, and more links, to move. We're honest about that with every client: head terms in a crowded market are a long game.

What's also true: several AI-style long questions, like "best data providers for b2b tam modeling", already rank in the top three. These get few clicks today, but they are exactly the kind of questions people now ask AI assistants.

What changed for the team

Generect went from a product site with no search presence to a blog that brings in developers and sales teams every day. The traffic lands on articles that end with the product, and on landing pages built for specific industries.

Want to build a blog that ranks this fast? Book a call.