A surveyor rings a Manchester terrace after a winter of hairline cracks along the party wall. The homeowner already called two firms from Google. They are not shopping “best foundation repair company.” They are comparing who can attend this week, who will write a clear scope for underpinning or crack stitching, and who will not vanish after the first quote. That is the search intent your PPC has to catch - and the reason a generic home-services account structure burns pounds before a single booked job appears.
This piece is a self-audit for ppc for foundation repair companies uk: score the account, find the leaks that inflate CPL while booked jobs stay flat, then fix in priority order. Currency is GBP, VAT sits on invoices where it always has, and the buyer language is British (estate agents, managing agents, council stock, terraced housing stock in Leeds, Birmingham, and Glasgow as much as London). The metric that matters is cost per booked survey or remedial job, not form volume.
Score the account before you touch bids
Most foundation repair Google Ads accounts in the UK look “fine” in the dashboard: impressions, a few hundred clicks, a CPL the owner can live with for a month. Then the diary is empty on Fridays because half the leads were DIY crack-filler shoppers, insurance claim fishing, or people 40 miles outside your real travel radius. An audit starts with economics and intent, not with another keyword idea.
Walk the last 60-90 days with the same questions you would ask a new survey crew: what did we pay, what did we book, what did we write off. Pull Search terms, call recordings or call-tracking logs, form submissions, and the diary of paid surveys. If you cannot join a click to a booked appointment within a few days, you are auditing noise.
Audit scorecard
-
Booked-job unit economics
Define the number that runs the business: cost per booked structural survey, cost per accepted underpinning or helical pile job, not cost per “lead.” Score yourself 0 if you only report form fills. Score high only if Google Ads spend ties to diary entries and written quotes in GBP, including VAT treatment you actually use with homeowners and commercial clients.
-
High-intent query coverage
Map spend to clusters that buy work: subsidence survey, underpinning quotes, foundation crack repair, wall tie failure, heave, party wall movement after nearby works. Soft DIY and “how to fill cracks in plaster” traffic should be near zero. If brand and competitor terms dominate while non-brand structural intent is thin, the account is under-serving the jobs that pay for vans.
-
Negative keyword hygiene
In our experience auditing UK trade accounts, a fifth or more of spend commonly leaks to irrelevant queries when negatives go unmined for more than a fortnight. For foundation repair, negatives usually need DIY, training courses, jobs/careers, free templates, insurance claim only, waterproof paint, damp proof injection sold as a cure-all when you do not offer it, and postcodes you never service. Score low if Search terms have not been mined weekly for 30 days.
-
Call vs form path match
Emergency and same-week structural worry converts on the phone. Planned quotes often use forms. If every ad routes only to a long form, or only to a generic number with no call reporting, you will misread quality. Score high when call extensions, click-to-call on mobile, and form paths are both measured and staffed to the hours you bid.
-
Landing page message match
Ad copy that promises “subsidence survey this week - Manchester & surrounds” must land on a page that repeats that promise, shows method photos, insurance and accreditation proof, and a single primary CTA. Homepages with six trades and a stock crack photo fail this row. Mobile is the majority of paid clicks in most markets; if the page is slow or the phone button is buried, mark this red.
-
Geo and radius honesty
Foundation crews do not magically cover the whole UK from one depot. Score the account on presence-targeting, radius by depot (London M25 vs West Midlands vs Central Belt), and exclusions for islands or counties you never attend. Impression share outside real travel time is a vanity metric.
-
Dayparting vs response capacity
Bids that run 24/7 while the office only answers 08:00-17:00 create expensive voicemail. Score high when ad schedules mirror estimator availability, emergency rota if you offer it, and weekend rules you can actually honour. Speed-to-lead on structural fear is measured in minutes, not “we’ll call tomorrow.”
-
Conversion tracking integrity
Smart bidding only helps when conversions mean booked surveys or qualified calls, not thank-you page hits from tyre-kickers. Enhanced Conversions and clean call conversions matter more as privacy and ad blockers eat signal. Score low if GA4, Google Ads, and the CRM or job book disagree by more than a small gap you can explain.
-
Creative and RSA proof load
Creative and RSA proof load RSAs need more than “trusted local experts.” Pin or heavily weight assets that name methods (underpinning, resin injection where you offer it, crack monitoring), accreditations, and response promise. Score the library: photos of real UK housing stock, before/after of stitching, surveyor on site - not generic US basement stock.
-
Budget waste and structure bloat
Too many micro-campaigns reset learning and hide which intent cluster pays. Fewer campaigns with tight themes, shared negatives, and clear labels usually beat a maze of experiments nobody reviews. Mark this row if you cannot explain each campaign’s job in one sentence tied to booked work.
Give each row a simple Red / Amber / Green. Anything Red on economics, negatives, tracking, or response capacity outranks a CPC optimisation. Directionally, search CPCs have been climbing year on year across categories; foundation and structural queries in dense UK cities sit at the expensive end of home services. You do not win by being the cheapest click. You win by refusing clicks that never become site visits that book.
Where UK foundation repair PPC quietly bleeds budget
The failure mode is specific. Someone searches after noticing diagonal cracking above a lintel or doors that stick after a dry spell. They click three ads. Two send them to a national homepage with a postcode widget and a promise of “free quote.” One lands on a page that names the failure mode, shows a local case, and offers a paid survey with a clear next step. Guess which diary fills.
Wasted spend clusters in predictable places. Broad match without ruthless negatives pulls searches like “crack filler homebase”, “foundation course CITB”, and “damp proof injection cost”, plus general builder enquiries that never need a structural survey. Even a tidy Search setup can still bleed if nobody owns the weekly prune and Search terms reports fill with that same junk. Agencies and in-house teams both fall into the trap of celebrating lead count while estimators complain the pipeline is soft.
Landing pages finish the damage. A slow mobile page, a form that asks for ten fields before a phone number, no Trustpilot or Checkatrade-style proof near the CTA, and no mention of PI insurance or method statements will convert fear into bounce. When traffic fails after the click, session behaviour usually shows the scroll dying before proof, or rage taps on a non-clickable phone number. Tools like HeyLead Insights help you see that pattern on the live page instead of guessing from bounce rate alone.
Response capacity is the silent partner of bid strategy. If Google Ads spends heavily after 19:00 and nobody answers, you paid for someone else’s morning callback. If an estate agent or managing agent needs a survey for a sale chain and you take 11 hours to reply, the job is gone. Audit call answer rates and first-response times on paid leads the same way you audit Quality Score.
If your scorecard is mostly Amber and Red on match and response, a structured SEM / Google Ads rebuild beats another month of “optimising” the same leaky funnel.
Get a free marketing audit - we review your search, ads, and landing pages and send back what to fix first.
Get a free audit

Checklist playbook: rebuild Search so it books surveys
A typical UK foundation repair account needs three to four weeks to stabilise after a structural rebuild - here is the order that protects learning while removing waste.
Action checklist
- Export 90 days of Search terms. Tag each row: bookable structural intent, adjacent (damp, drainage that you may or may not take), junk. Build negative lists from junk immediately. Add phrase and exact themes for subsidence survey, underpinning, foundation crack repair, wall ties, heave, and local modifiers you truly cover.
- Split campaigns by intent and depot, not by every synonym. Example: Survey/diagnosis vs Remedial methods vs Brand. Keep location assets honest per depot (e.g. Birmingham & Black Country vs Leeds corridor). Pause anything outside drive time you will not serve.
- Rewrite RSAs around the job, not the slogan. Headlines that name the problem and the offer (“Subsidence survey this week”, “Underpinning quotes - clear scope”) outperform vague trust claims. Include path fields that mirror the landing H1.
- Build one dedicated landing per major intent cluster. Above the fold: problem recognition, what you inspect, CTA for call and short form. Mid-page: method photos from UK stock, accreditations, typical timeline, VAT-aware pricing posture if you publish bands, and what happens after the survey. Keep secondary links quiet.
- Instrument calls and forms as separate conversions with sensible values. A booked survey is worth more than a brochure download. Use call tracking numbers on ads and landings so estimators can score quality. Feed only qualified outcomes into smart bidding once volume is stable.
- Align ad schedule to phones answered. Tighten evening and Sunday bids unless a duty estimator exists. Raise bids on weekday mornings when managing agents and homeowners both tend to act on survey need.
- Fix mobile speed and INP on the landing. Core Web Vitals still separate usable pages from ones that drop paid traffic; many origins still fail the full set. Compress hero images of cracks and scaffolding, defer non-essential scripts, keep the click-to-call sticky.
- Add proof blocks that reduce fear: sample report redacted, insurance wording summary, reviews that mention punctual surveys and tidy sites, photos of monitoring gauges if you use them. Proof near the form beats a buried footer badge.
- Set a weekly operating loop: Search terms prune, RSA asset report, landing conversion rate by device, cost per booked survey from the job book. Do not reshuffle structure daily; constant resets punish learning.
- Only then test bid strategies. Manual CPC or a conservative tCPA target on thin data often thrashes the learning phase without improving quality - define clean conversions first, then let automation work with real signal. When signal is solid, automation can help - it will not invent quality you never defined.
Free tools - try these yourself
Two operational notes that keep UK foundation accounts honest. First, insurance-led work and cash private work behave differently; if ads promise “insurance claims handled” you must staff that path or you will attract files you cannot progress. Second, competitor conquesting can look cheap until you measure booked rate - many clicks are comparison shoppers who already have a preferred engineer.
For operators who want channel and niche context in one place, HeyLead’s Foundation Repair marketing overview sits alongside the paid search work above without turning this checklist into a full-funnel lecture.
Two UK scenarios: what broke and what fixed booked-job maths
West Midlands remedial firm, multi-depot ads, one homepage. Spend sat near the top of their comfort band. CPL looked acceptable. Estimators said most “leads” wanted general builders or free advice on hairline plaster. Search terms were full of DIY and damp. The fix was not a higher budget. They carved a Survey campaign with exact and phrase structural themes, applied a thick negative list, and launched a mobile landing with click-to-call, three local job photos, and a short form asking only address, crack description, and preferred call window. Call conversions were valued higher than forms. Within a few weeks the share of diary-ready surveys rose sharply; cost per booked survey fell even though CPC on the remaining terms stayed high. The mechanism was intent isolation plus answerable phone path, not a clever bid strategy. The same pattern shows up when teams celebrate a low CPL until the survey diary reveals half the forms were paint-over advice - negatives and a survey-only landing fix the maths faster than any bid change.
Scottish belt specialist bidding overnight on brand-plus-generic. Ads ran 24 hours. Voicemail filled after 18:00. Smart bidding chased volume into cheap hours. They cut schedule to staffed blocks, added call-only emphasis on mobile for emergency-adjacent queries they truly covered, and required same-morning callback SLAs on paid leads. Booked rate on answered calls moved from a weak fraction to a workable majority. Spend dropped slightly; revenue from paid did not, because unpaid overnight clicks stopped masquerading as demand. Overnight bids can look efficient in Google Ads while you are really paying for voicemail; once the schedule matches the estimator rota, cost per booked job often drops without touching keywords.
Neither story needed a dozen platforms. Both needed the audit rows that owners skip when the interface looks busy.

FAQs
What should I optimise first in a foundation repair Google Ads account?
Start with Search terms and negatives, conversion definitions tied to booked surveys, and landing match. Bid strategy changes on dirty data usually raise spend without raising diary quality.
Is CPL a useful KPI for UK foundation repair PPC?
Only as a triage number. The operating KPI is cost per booked survey or accepted remedial job in GBP, after you filter DIY and out-of-area enquiries. A rising CPL with rising booked rate can still be healthy if ticket sizes support it.
Should we use Performance Max for foundation repair?
Only after Search is clean and conversion signals are trustworthy. PMax can find volume; without tight assets, strong landings, and negatives discipline elsewhere in the account, it often rediscovers the same low-intent traffic you already blocked in Search.
How fast do we need to answer paid leads?
Treat structural fear and sale-chain surveys as near-real-time. Same-hour response on phone leads is the practical bar; next-day email-only habits lose jobs to the firm that picks up.
Do we need separate landings for every city?
Not always. Separate by intent and depot capacity first. City pages help when proof, travel time, and phone routing truly differ - cosmetic city swaps without local proof rarely move conversion rate.
Putting it to work
This week
-
Score all ten audit rows Red / Amber / Green with 60-90 days of data open beside the diary.
-
Export Search terms; add a first-pass negative list for DIY, careers, pure damp curiosity you do not serve, and out-of-area postcodes.
-
Verify call and form conversions fire and that thank-you events are not the only “conversions” smart bidding sees.
-
Run your main paid landing through the Core Web Vitals checker and fix the phone CTA on mobile if it is below the fold.
-
Align ad schedule to hours a human answers; pause pure overnight if nobody is on rota.
-
Tag every new paid URL with the UTM link builder so source joins the job book cleanly.
Next 30 days
-
Ship one intent-matched landing for your highest-spend structural theme with UK proof and dual call/form paths.
-
Collapse noisy campaign clutter into a few intent themes with shared negatives.
-
Review RSA assets weekly; replace vague trust lines with method and response promises.
-
Report cost per booked survey to ownership, not lead count alone.
-
If post-click behaviour is still opaque, use session insight on the landing to find where proof and forms leak before you raise budgets.
Pull your last 60 days of Google Ads search terms against the survey diary and mark every pound that never had a chance to book. If you want a partner to own the messy middle - negatives discipline, call-quality conversions, landing match, and schedules that respect real estimator capacity for foundation repair PPC in the UK - HeyLead can run that programme end to end. Start a conversation on Chat with us on WhatsApp.
Free marketing audit, or reach Martin directly:
Get a free audit Chat with us on WhatsApp · martin@heylead.com