Water treatment Google Ads audit checklist for UK firms

The call log on a Monday tells you more about your Google Ads than the dashboard does. Softener installs booked from “hard water near me”. A borehole pump quote that never got a site visit because the van was already in Leeds. Three “water filter” form fills that turned out to be students researching science projects. If you sell water treatment, well pumps, UV systems, or borehole work across the UK, paid search is where urgent, high-intent demand still shows up first - and where budget quietly leaks when structure, negatives, landing pages, and response capacity are out of sync.

This piece is a self-audit. Score the account the way an operator would: against booked jobs in GBP, not vanity CPL. Work through the checklist, mark what is broken, then run the 30-day sprint at the end. It is written for owners, founders, and marketing leads running PPC for water treatment and well pumps companies UK who need fewer tyre-kickers and more surveys that become installs.

What a good paid-search week looks like for UK water and well firms

UK buyers rarely “browse” water treatment. A hard-water complaint after a boiler service, a private supply failing after heavy rain, a commercial site that needs UV before a hygiene audit, a landlord chasing a filter quote before a tenancy change - those are the moments that hit Search. Intent is local, often same-week, and split between homeowners, estate agents chasing property standards, facilities managers, and agricultural or rural properties with boreholes.

A good week is not “more leads”. It is booked site surveys and installs at a cost you can defend. That means search terms that map to real jobs (water softener installation, iron filter, UV disinfection, borehole pump replacement, well pump repair), call tracking that survives after-hours, and a landing page that proves you actually do the work in their postcode area - not a generic home-services homepage with a stock tank photo.

Capacity matters as much as CPC. If your engineers are already booked solid Thursday to Saturday across Manchester and Birmingham, dayparting and bid adjustments that still push Friday night traffic will buy you expensive voicemail. Audit Google Ads and ops together or you will “win” auctions you cannot answer.

Mobile drives most of the clicks. Conversion still often finishes on a call. If your account only optimises for form submits, you are training bidding on the wrong half of the journey. Keep that in mind as you score the sections below.

Where UK water treatment PPC budgets actually leak

Across paid search, 20-30% of spend routinely dies on irrelevant queries, weak negatives, or landing pages that do not match the ad. Water and well work is worse than average because the language is shared with DIY, education, and wholesale parts. “Water filter”, “well pump”, and “UV water” pull students, aquarium hobbyists, spare-part shoppers, and people looking for council advice - not always a survey.

Another leak is metric choice. CPL looks healthy while cost per booked job climbs. A £40 form fill that never books a survey is not cheaper than a £90 call that books a softener install at £1,200+. If sales and the diary are not closed-loop with the account, Smart Bidding chases the wrong conversion.

Landing mismatch finishes the job. Ads promise “same-week softener survey in Glasgow”; the click lands on a national brochure site with one contact form and no proof of WaterSafe-style competence, insurance, or local case photos. Operators who only glance at last-click ROAS miss that. Platform ROAS benchmarks for home services vary widely and rarely account for whether your conversion is a soft enquiry or a booked job - which is why cost per booked survey is the only number that matters here.

If you already know traffic is arriving and still not converting, pressure-test the page with behaviour data - scroll depth, rage clicks on phone numbers, form abandon - before you raise bids again. Tools like HeyLead Insights make those leaks visible without guessing.

For a partner that stays on the SEM side of this niche rather than a one-off media plan, see how we approach SEM / Google Ads for UK home-service firms.

Google Ads audit scorecard for water treatment and well pumps

Work the scorecard in order. Score each item honestly: pass, partial, or fail. Failures that touch search terms, conversion definition, or response speed outrank pretty RSAs. This is the core of a practical audit for ppc for water treatment and well pumps companies uk.

Audit scorecard

1. Intent map vs real jobs List the last 30 booked jobs by type: domestic softener install, salt delivery only, commercial UV, borehole pump swap, well head repair, iron/manganese filter, UV + sediment package, emergency no-water callout. Pull Search Terms for 90 days and tag each query as bookable, research, parts/DIY, wrong trade, or geo-mismatch. If more than about a fifth of spend sits outside bookable and adjacent commercial intent, structure is wrong before creative is. Separate campaigns or ad groups for emergency pump failure versus planned softener quotes. Do not let “water treatment” as a single theme buy everything from lab equipment to fish tanks.

2. Negative keyword discipline Build and maintain negatives for jobs you do not want: DIY, how to, YouTube, PDF, salary, course, apprenticeship, aquarium, fish tank, car wash, swimming pool (unless you truly serve pools), wholesale, spare parts only, second hand, Gumtree, eBay, council grant research with no commercial intent, brand names you do not install. Add local noise: job titles, “meaning of”, and pure definitional queries. Review search terms weekly for the first month of any restructure, then at least fortnightly. Negatives are not a one-time list; UK query mix shifts with weather, boiler service season, and rural water quality stories in the press.

3. Match types and brand protection Protect exact and phrase on high-value install and repair phrases. Use broader match only where conversion tracking is clean and negatives are mature. Brand campaigns should catch your company name and common misspellings so competitors cannot sit on them cheaply. If Performance Max is “driving the bulk”, open asset group search insights and force the same negative discipline - PMax will happily explore soft educational queries if you let it.

4. Geo and radius reality Presence targeting and radius must match where vans actually go. A London HQ that “covers the South East” on the website but only profits inside the M25 will burn budget on coastal and rural clicks you never survey. Use location report: user location versus location of interest. Exclude postcodes you systematically decline. For multi-depot setups (e.g. Midlands plus North West), split campaigns so bids and ad copy name the right city set - Manchester, Birmingham, Leeds - rather than one UK-wide soup.

5. Ad copy proof and offer clarity RSAs should lead with the job and the proof: WaterSafe or equivalent membership if held, insurance, years on boreholes, same-week survey, free water test where you genuinely offer it, commercial compliance language for B2B. Pin at least one headline that states the service and one that states the area. Avoid vague “quality water solutions”. Extensions: call, structured snippets (softeners, UV, borehole pumps, iron filters), price where honest, and sitelinks to dedicated service pages - not blog posts.

6. Call vs form conversion design Import calls from call tracking with a minimum duration that approximates a real conversation (often 60-90 seconds, tune to your script). Count form submits only when they include postcode, property type, and problem type - or route soft forms out of primary Smart Bidding. Enhanced Conversions and clean GTM firing matter when privacy tools strip 20%+ of browser signals. If you only optimise for “submit”, expect students and tire-kickers to look like success.

7. Landing page match and speed One primary URL per intent cluster: softener install, UV/commercial, borehole/well pump repair. Above the fold: service, area, phone, and one proof block (photo of real install, review snippet, accreditation). Forms short; click-to-call sticky on mobile. Check Core Web Vitals - fewer than half of origins pass all three Core Web Vitals thresholds, and slow mobile pages punish paid traffic harder than organic. Use dedicated pages, not the homepage. If someone says they “just send traffic to your homepage,” treat that as a red flag.

8. Dayparting and response capacity Map when surveys actually get booked. If phones die after 17:30 and Saturdays are admin-only, cut or lower bids then unless you have a vetted answering service that books diaries, not just takes messages. Align ad schedules with engineer dispatch reality across your densest corridors. Response speed is a conversion factor: a softener lead answered in under five minutes books differently from one returned next morning.

9. Bidding and learning hygiene New accounts under conversion pressure often need manual or constrained bidding until 30+ quality conversions exist per goal. Automated bidding without clean goals inflates spend without improving job quality - a complaint you will hear in every PPC forum for a reason. Avoid daily restructures that reset learning. When you change goals from form to qualified call + booked survey, expect a relearning window and protect brand and exact install terms meanwhile.

10. Cost per booked job reporting Build a simple weekly sheet: spend (GBP, ex VAT clarity), qualified enquiries, surveys booked, installs won, revenue, cost per booked survey, cost per install. Tie UTM and call source through to the job ID in your CRM or even a disciplined spreadsheet. Last-click will double-count in multi-touch journeys; still, directional cost per booked job beats shiny platform CPL. Reject any promise of a fixed ROAS or CPL before someone has seen your account, pages, and unit economics.

DIY playbook: run the audit in one working session

1. Export last 90 days of search terms, locations, and auction insights. Keep raw CSVs.

2. Tag every term against your real job list; sum spend on non-bookable tags.

3. Write a negative list from those tags and apply at campaign level where safe.

4. Split or label campaigns so emergency pump work is not funding soft research traffic.

5. Verify call and form tags in GTM; test from a mobile on 4G, not office Wi-Fi only.

6. Open each ad’s final URL and check message match in 10 seconds on a phone.

7. Align ad schedule with when someone who can book a survey actually answers.

8. Define one primary conversion for bidding: qualified call or qualified form, not both equal.

9. Build the cost-per-booked-job sheet and fill two prior weeks so you have a baseline.

10. List only three fixes for the next fortnight - usually negatives, landing match, and call tracking.

Two UK account snapshots where one audit fix changed bookings

A family-run softener and filter firm covering Bristol and the surrounding counties was happy with a £38 CPL. Diary interviews showed most forms were “how much does a softener cost” with no postcode and no water hardness context. Search terms were heavy on generic “water softener” plus a surprising volume of “best water softener 2026” listicle traffic. They added aggressive negatives for pure research phrasing, moved exact match onto “water softener installation” plus town modifiers, and rebuilt the landing page with a three-field form (postcode, homes vs commercial, existing softener yes/no) and click-to-call. Primary optimisation shifted to calls over 75 seconds plus forms with postcode. Booked surveys per £1,000 spend rose within three weeks; CPL went up slightly and nobody cared because installs followed.

A borehole and well pump specialist near Edinburgh ran a single UK-wide Search campaign. Location reports showed expensive clicks from regions they never travel to. They cut geo to profitable drive-time bands, wrote RSA headlines that named rural private supply and pump failure, and stopped sending emergency queries to a slow multi-service homepage. They also stopped bidding hard after 18:00 when only a mobile went to voicemail. Cost per booked callout dropped even though average CPC rose - fewer irrelevant miles, faster answer rate on the clicks that remained.

If your niche page story needs a wider marketing frame beyond this audit, the industry overview on Water Treatment and Well Pumps marketing sits alongside the channel work above.

FAQs

How often should a UK water treatment Google Ads account be audited?

Run a full scorecard quarterly, and a search-term plus negatives pass every one to two weeks while spend is active. After any landing page or tracking change, re-check conversion actions within 48 hours so Smart Bidding is not trained on broken goals.

Should we use Performance Max for softeners and well pumps?

Only after Search is clean: solid negatives, dedicated landers, and a conversion definition tied to booked work. If PMax is already live, audit search themes and exclusions the same way you would a broad match campaign. Do not let PMax become a second ungoverned budget. In this niche, PMax without tight asset-group themes and a solid exclusion list will drift into aquarium, DIY, and educational placements faster than most verticals - audit search themes in the Insights tab monthly, not quarterly.

What is a sensible primary KPI - CPL or cost per booked job?

Cost per booked survey or cost per install in GBP. CPL is a diagnostic, not the scoreboard. Pair it with answer rate and travel fit so cheap leads from unreachable postcodes do not look like wins.

Do we need separate campaigns for commercial UV and domestic softeners?

Usually yes. Message, proof, ticket size, and sales cycle differ. Separate campaigns keep budgets and bids from subsidising the wrong buyer and make RSA testing clearer.

Where do landing pages fit in a PPC audit?

They are part of the account. Message match, mobile speed, proof, and form friction decide whether qualified clicks become jobs. Behaviour tools help you see abandon points; fix those before scaling bids.

Putting it to work

This week

• Export 90 days of search terms and tag spend into bookable vs waste buckets for your real job types.

• Ship a first negative list and remove or pause obvious non-service themes.

• Test call tracking and form tags from a mobile handset; fix broken fires before touching bids.

• Pick the top ad by spend and rebuild its landing page match in one pass (headline, proof, postcode field, click-to-call).

• Set ad schedule to hours when a human can book a survey.

• Open a simple cost-per-booked-job sheet and baseline the last two weeks.

Next 30 days

• Split emergency pump intent from planned treatment quotes if they still share one budget.

• Tighten geos to profitable travel bands; exclude habitual no-go postcodes.

• Move Smart Bidding primary conversion to qualified calls or qualified forms only.

• Run one controlled RSA test on proof-led headlines (accreditation, local installs) versus generic claims.

• Review search terms fortnightly; add negatives and note new bookable phrases for exact coverage.

Start by pulling ninety days of search terms beside your last month of booked softener, UV, and pump jobs, then mark every pound that never could have become a survey. If you want that full loop - query hygiene, landing match, call tracking, and bidding aimed at cost per booked job rather than vanity CPL - HeyLead can run the ongoing Google Ads programme for UK water treatment and well pump firms so you are not re-auditing alone every quarter. Message us on WhatsApp or get in touch via our contact page.