What a teardown contains
A teardown costs £1,500 + VAT, takes 10 working days, and comes off your first month if you then hire PYC monthly. You pay before it starts, so here is exactly what you get, and one section worked through on real data.
Contents
- Summary. The three things to fix first, and why those three.
- How your search traffic is really moving. A like-for-like comparison of the same dates a year apart, searches for your name separated from everything else, the same searches compared at the same position, and how much of your traffic the data covers.
- Searches you don’t properly answer yet. Where your coverage of your subject is thin, both close to what you sell and in the questions around it.
- Pages dragging the site down. Pages Google shows but nobody clicks, thin pages, and pages competing with each other for the same searches.
- Internal links. How your pages link to each other compared with how the searches relate, and where that costs you.
- What sits above you in Google. AI Overviews, product carousels and local maps on your main searches, checked signed out, so the results are not shaped by anyone’s browsing.
- The roadmap. What to do, in order, with the effort and expected effect of each step, to run yourselves or hand to PYC.
- Method and data notes. Exactly what was measured, over which dates, and what the data cannot show.
Worked example: section 2, how search traffic is really moving
This is section 2 for one real account, an online signage shop, using figures already published in the click-through note. Search terms, names and locations are removed, as they are in anything PYC publishes.
What was compared
15 September to 31 October 2025 against the same dates in 2026. Both periods come after Google stopped serving 100 results per page in mid-September 2025, which removed many automated impressions from Search Console, and the matching dates rule out the normal seasonal pattern. The searches Google lists cover 35% of this account’s 2026 clicks, so the findings were also checked at page level, which covers nearly all of them.
The headline
Clicks rose from 1,271 to 1,562 and impressions from 85,444 to 115,133, but click-through fell from 1.49% to 1.36%. On its own that could mean the site is simply being shown for more searches. The test below separates that from people choosing the listing less.
| Position band | Searches | 2025 | 2026 | Average position |
|---|---|---|---|---|
| 1 to 3 | 60 | 2.84% | 2.11% | 2.5 → 3.3 |
| 4 to 10 | 291 | 0.85% | 0.69% | 6.6 → 6.8 |
The checks
- Positions 4 to 10 are the cleaner test, because average position barely moved there. In the top three it slipped, which could explain part of that fall.
- The fall in positions 4 to 10 holds for UK searches only, on desktop and on mobile.
- At page level, the 106 pages that stayed in positions 4 to 10 went from 1.31% to 1.02%.
- Most of the lost clicks are on product searches.
- Product carousels do not explain it: in positions 4 to 10, searches with and without them were clicked at about the same rate (0.67% and 0.63%).
- Visits from AI tools and Bing grew from 13 to 78, with AI tools bringing 8 orders, but that is small next to the roughly 116 Google clicks these searches lost.
What the finding means
Fewer people chose the listing at the same position, mostly on product searches, and the cause is not yet established. More rankings alone would not fix that. In a full teardown this feeds the roadmap: start with the searches in positions 4 to 10, look at what Google now shows above the listing, and test titles, descriptions and product data before building more pages.
Before you book
The teardown is yours to keep whether or not you hire PYC afterwards. Terms are on the terms page.