- ·Consolidate before expanding — cleanup precedes compounding in every turnaround.
- ·GBP is operational, not project work — build it into the shop workflow.
- ·Alignment is a deliverable; the founder conversation is where the plan gets shipped.
- ·The compounding phase begins around month 7 — structure the engagement to reach it.
Carage Auto is one of our longest running engagements. The story is worth telling because most of the wins came from unglamorous work — technical cleanup, information-architecture consolidation, and operationalised Google Business Profile management — not from anything you'd put on a conference slide.
This is the month-by-month reconstruction. Where we made mistakes, we name them. Where the compounding kicked in, we show what preceded it. If you're evaluating an agency or planning your own turnaround, this is the honest version of what a real engagement looks like.
The starting position
Solid business, weak digital footprint. Traffic flat for 14 months, near-zero map-pack visibility outside the immediate neighbourhood, a website that hadn't been touched structurally in years, and roughly 180 pages of legacy content — most of it thin, some of it duplicative, none of it clustered.
GBP was claimed but under-managed: primary category chosen years ago and never revisited, no posts, no Q&A management, photos from 2019, review response rate under 20%. The site had schema on a handful of pages, most of it Article schema on non-article content — Tier 3 in the sense of our schema post above.
The client's own frustration was the leading indicator: 'we're the best independent shop for European specialists in the metro and Google acts like we don't exist.' That was directionally correct. Fixing it took twelve months of disciplined work.
Month 1: the audit and the argument
The audit is not the deliverable. The argument is the deliverable. We spent as much time on the executive presentation as on the audit itself, because nothing ships without alignment — and in a founder-led business, alignment means a one-hour conversation where the founder sees the case for each priority and gets to push back before we write any code or copy.
The audit surfaced three priority buckets. Technical debt — canonicals, redirect chains, JS-rendered content that Google couldn't see, sitemap and robots hygiene. Information architecture — 180 pages of legacy content that needed to consolidate to roughly 100 pages organised into 6 clusters. GBP operational rhythm — from checkbox to channel.
The argument, in one sentence: 'we will consolidate before expanding, operationalise GBP as a weekly channel, and expect first meaningful ranking movement in month 4.' The founder said yes. Every subsequent decision came back to that sentence.
Months 2–3: technical debt and IA
A ruthless technical pass — canonicals rewritten, redirect chains flattened, JS-rendered content ported to server-rendered templates, sitemap regenerated dynamically from the CMS, robots.txt cleaned up. This work took about six weeks of developer time. It didn't move traffic, and it wasn't supposed to. It removed the technical drag that would have blunted every subsequent gain.
The IA redesign was harder because it was political. Consolidating 180 pages to 100 meant retiring pages that had sentimental value to internal stakeholders. We handled this with a scoring pass — each page scored on traffic, conversion, and topical fit — and let the numbers do the arguing. Pages scoring below threshold were consolidated or 301'd, with the surviving pages inheriting the collective link equity.
By end of month 3, the site was 40% smaller by page count and about 15% larger by average page depth and quality. Traffic was flat. That was fine — we hadn't shipped any content yet.
Months 4–6: content and GBP
Service clusters for each core repair category: European diagnostics, timing belt and chain, cooling system, brakes, transmission, pre-purchase inspection. Each cluster was a pillar page plus 6–10 supporting pages, built in waves with an internal-link pass after each wave, on the schedule described in our clusters post.
GBP got operationalised weekly. Posts on Monday, Q&A audit Tuesday, review outreach built into the shop workflow (technicians asked at pickup, with a texted review link), competitor pack scan Thursday, weekly report Friday. Reviews grew from 4/month to 12/month within eight weeks and stayed there.
First map-pack wins showed up in month 5, faster than we'd promised. By end of month 6, the shop was in the top-3 for its primary service qualifier across the neighbourhood and adjacent neighbourhoods, and the phone volume was up roughly 35% on the same marketing spend.
Months 7+: the compounding phase
This is the part clients love and agencies rarely reach. Traffic began compounding — the same cluster earning new keywords quarter after quarter without new pages, because the internal linking density kept improving and the republishing cadence kept the pages fresh.
The interesting compounding wasn't the direct queries; it was the long-tail. The European diagnostics cluster started ranking for specific vehicle-and-symptom queries ('BMW N54 misfire diagnostic Austin', 'Mercedes W211 airmatic compressor Austin') that we had never explicitly targeted. Those queries convert at 6–8× the rate of head terms because the intent is so specific.
By month 12, organic traffic was up 214%, GBP-attributed calls up 340%, and the shop had a six-week booking backlog for its highest-margin services. The engagement is ongoing; the cluster set has since expanded to cover pre-purchase inspection and fleet service, both of which are compounding on the same pattern.
The mistakes worth naming
We under-scoped the IA politics. We treated page consolidation as a technical exercise; it was 40% technical and 60% stakeholder management. Next time we build a two-week buffer for the consolidation conversation before we start executing it.
We were slow to add Person schema to the shop's lead technician for the diagnostic pages. When we finally did, in month 8, we saw a modest lift in citation and AI Overview inclusion. Should have done it in month 2.
We over-invested in a couple of vanity clusters that turned out not to have search demand behind them. Killed those in month 6 and reallocated the budget. The lesson is that even a good process needs to interrogate its assumptions quarterly.
Lessons transferable to other verticals
Consolidation beats expansion. If you have 200+ legacy pages and flat traffic, cutting is almost always the right first move. New content on top of a bloated site earns less than new content on top of a cleaned site, and the cleanup takes weeks, not quarters.
GBP is operational, not project work. The businesses that hold local rankings do this weekly. The ones that treat it as a quarterly refresh lose ground to the ones that don't.
Alignment is a deliverable. The best audit in the world doesn't ship if the founder disagrees with a priority. Front-load that conversation and be willing to change the plan if the founder surfaces something you missed.
The compounding is real but it lives past month six. Most engagements that fail, fail because either the client or the agency loses patience before the compounding phase begins. Structure the engagement so both sides can afford to reach month seven.
We ship this exact playbook in engagements.
If what you just read maps to a problem in front of you, a 30-minute conversation is the fastest way to know if we can help.