Eight months of decline, reversed in one.
Included because reversing a decline is a harder problem than growing a healthy site — and it is the situation most brands actually call us in for. A headless Drupal stack with a Next.js front end, an eight-month negative trendline, and no internal agreement on the cause.
Four numbers. One month.
The trendline had been negative for eight months before we started. It reversed in one. This is what a diagnosis-first engagement looks like.
Eight consecutive months of organic traffic decline before engagement began — with no internal consensus on the root cause.
Source · Client GSC & GA4 · Pre-engagement baselineTraffic recovery in month one following diagnosis and implementation. The trendline reversed before month two began.
Source · Client GSC & GA4 · Month-one resultOne month from diagnosis to trendline reversal — because the diagnosis was correct and the fix was executed, not debated.
Source · Client GSC & GA4 · Engagement timelineRender and indexation, internal linking restructure, and content refresh — run in parallel, not sequentially.
Source · Internal · Engagement workstream logEvery team had a theory that exonerated it.
A headless Drupal site front-ended with Next.js, an eight-month negative trendline, and an internal standoff between engineering, editorial, and SEO — each with a competing diagnosis, none of which pointed at their own layer.
The engagement started in a familiar place: a large organisation with a declining trendline and a split internal diagnosis. Engineering pointed at algorithm updates. Editorial pointed at technical issues. SEO pointed at content quality. Each theory was defensible. None of them was moving the needle.
On a headless stack — Drupal as the CMS, Next.js as the front end — the render pipeline is the layer that has no natural internal advocate. It sits between engineering and SEO in a way that makes it easy for both teams to assume the other has it covered. On this site, nobody did.
The render and indexation audit was the first thing we ran. It was mostly the first two problems: render and internal linking. The content refresh came third — and only after the first two were fixed, because refreshing content that search engines can't render correctly is effort that lands nowhere.
- Stack — Headless Drupal · Next.js front end
- Trendline — 8 months consecutive decline
- Internal diagnosis — Three competing theories, no consensus
- Root cause — Render pipeline + internal linking
- Content refresh — Third priority, after render fixed
Three workstreams. One month.
Diagnosis first, then execution across all three workstreams simultaneously. The order mattered: fixing render before touching content meant every content change actually landed.
- 01Workstream One
Render and indexation diagnosis
On a headless stack, the render layer is where problems hide — and where they were. A full crawl-and-render audit across the Next.js front end identified which pages were being delivered to Googlebot fully rendered versus which were being served as empty shells that JavaScript never filled for the crawler. Indexation coverage was mapped against GSC data to identify the gap between pages the site thought were indexed and pages Google was actually reading.
- 02Workstream Two
Internal linking restructure
Equity was not being routed into the pages that were supposed to be earning it. A full internal link audit mapped the existing flow against the site's commercial and informational priorities. The restructure re-pointed internal links toward the highest-value pages — mutual fund scheme pages, category hubs, and high-intent calculators — and removed equity leaks into low-priority pages that were absorbing link weight without returning traffic.
- 03Workstream Three
Content refresh sprint
Aimed at the pages that had slipped furthest, fastest — identified by cross-referencing GSC position data with traffic loss over the decline period. Content refreshes were prioritised by pages where position loss was consistent rather than algorithmic, meaning the page had the right intent match but had been overtaken by fresher, more authoritative content. These were updated after render was fixed, so the refreshed content actually reached Google.
- Render audit — full crawl-and-render diagnosis across the Next.js front end to identify what Googlebot was and wasn't reading.
- Indexation mapping — GSC coverage analysis to close the gap between assumed and actual indexed pages.
- Internal link restructure — equity re-routed into high-value scheme pages, category hubs, and intent-matched calculators.
- Equity leak removal — low-priority pages absorbing link weight identified and deprioritised in the link graph.
- Content refresh sprint — pages with consistent position loss updated after render was confirmed working, so changes actually landed.
Every team had a theory that exonerated their own layer. The render pipeline had no advocate.The Bee Online diagnosis · Tata Mutual Fund recovery engagement
Built for enterprise sites where the problem has no internal owner.
What we ran for this engagement wasn't a standard SEO retainer. It was a diagnosis-first recovery — starting from the render layer because that is where headless stacks hide problems that look, from the inside, like algorithm updates or content quality issues.
If your site has been declining for months with no clear internal consensus on why — especially on a headless or JavaScript-heavy stack — the same approach applies. The fix is usually not more content. It is finding what search engines can't read, and fixing it before anything else moves.
A simplified, prospect-safe view of a recovery engagement. Specific tactics, tools, vendors, and technical implementation detail are withheld by design.
Shared with permission. Not for redistribution. Specific strategies, vendor stack, and technical detail available under NDA.