Digital Marketing Singapore
SEO & Lead Generation Agency
Case Study | Keyword Research

Waterproofing Contractor Keyword Mapping Singapore: Organic Enquiries Up 229% in 6 Months

Client Type
Waterproofing and leak repair contractor, about 25 staff, homes and small commercial buildings across Singapore
Services
Keyword research and mapping, then 14 new pages built from the map
Engagement
6 months
Primary Result
Organic enquiries from 7 to 23 a month
386Keywords mapped into 4 intent groups
+229%Monthly organic enquiries
58Keywords on Google page 1, up from 6

Client name withheld under NDA.

The Challenge

The Starting Point

Waterproofing is a trade where the owner and the customer use completely different words. Contractors talk about torch-on membranes, PU injection and cementitious coatings. Homeowners talk about a brown stain on the ceiling, a toilet that leaks into the flat below, or a wall that turns damp every time it rains. Searches for waterproofing singapore and waterproofing contractor singapore are competitive, and most of the people with an actual leak never type either phrase. They search for the symptom, often late at night and often while worrying about who will pay for it.

This contractor had built a good business on referrals and repeat work from property managers. Its website had 9 pages, and every one of them was named after a method: membrane waterproofing, PU injection, roof coating, and so on. Google showed the site for 6 keywords on page 1, almost all of them variations of the company's own service names. The site drew about 640 organic visitors a month and produced around 7 enquiries. The owner asked us a simple question: what are people actually searching for when they need us, and which pages should we build to be there?

MetricMonth 0
Keywords mapped to a page0
Pages built around a keyword group9
Keywords on Google page 16
Monthly organic visitors640
Share of organic visits landing on leak problem pages0%
Monthly organic enquiries7
What We Did

What We Did

Phase 1, Month 1

Collecting the Words Customers Actually Use

We started with the client's own records, not a keyword tool. The office manager exported six months of WhatsApp messages and web form enquiries, 212 in all, and we noted the exact words people used to describe their problem. Then we pulled search data from SEMrush, Google Search Console, autocomplete and the People Also Ask boxes. That gave us 1,140 phrases, which we filtered down to 386 that described a real waterproofing or leak problem in Singapore homes and buildings.

Starting with real messages kept the research honest. Keyword tools show volume, not how a worried homeowner describes a leak. Almost nobody asked for waterproofing by name.

Phase 2, Months 1 to 2

Grouping by Intent and Finding the Surprises

We sorted the 386 keywords into four intent groups. Problem searches, like a ceiling leaking after rain, made up 171. Responsibility searches, about who pays when a leak comes from the flat above, made up 64. Method searches, which used the contractor's own terms, made up 92. Hiring searches, like roof leak repair singapore or a contractor near a specific estate, made up 59. Three things surprised the owner. The word seepage, which nobody in the company used, appeared in 41 of the mapped keywords. The responsibility group had no page anywhere on the site. And PU injection, the service with the best margin, had almost no searches by that name, while searches about a cracked wall leaking in heavy rain described exactly the job it fixes.

Intent matters more than volume here. A person searching for who pays for a ceiling leak in an HDB flat is close to calling someone, even though the phrase looks like a question. Grouping by intent showed the client which searches were worth a page and which were only worth a line in an answer.

Phase 3, Month 2

Building the Keyword Map

We turned the groups into a keyword mapping sheet with one row per page. Each page got one primary keyword, three to eight supporting keywords, the intent it served and the question it had to answer in its first paragraph. Five of the existing method pages were kept but re-pointed at the problem a customer would search for, so the PU injection page became the page for walls that leak when it rains. Fourteen new pages were planned. Keywords that overlapped were assigned to a single page so the site would not compete with itself.

A keyword list on its own does not change anything. The map is what tells a writer what each page is for and stops two pages chasing the same search. It also gave the owner a way to say no to page ideas that had no searches behind them.

Phase 4, Months 3 to 4

Writing the 14 Pages From the Map

We wrote and published the 14 new pages over eight weeks: eight problem pages (ceiling leaks, toilet leaks into the unit below, wall seepage, window leaks, flat roof ponding, planter box leaks, balcony leaks and landed house roof leaks), three responsibility guides for HDB flats, condos and landed homes, and three hiring pages, including one for ceiling leak repair singapore. Each page opened with what the customer was seeing, explained the likely cause in plain words, then showed the repair method and photos from the client's own jobs. We also tightened the on-page SEO on the five re-pointed pages.

The responsibility guides did something the client did not expect. They brought in condo residents and property managers who had not decided whether to call anyone yet, and the guides gave them a clear next step. Several of those readers became the larger, multi-unit jobs the contractor wanted.

Phase 5, Months 5 to 6

Tracking the Map and Filling the Gaps

We tracked all 386 keywords by page every week and checked which ones Google was matching to the wrong page. Eleven keywords were moved to a better page and the internal links were adjusted to match. Search Console also showed 23 new phrases the map had missed, mostly about leaks after renovation work, and we added them to existing pages.

A keyword map goes out of date the day it is finished. Checking it against real search data each month is what kept the new pages climbing instead of stalling at page two.

The Results

The Results

By Month 6 the site had 23 pages built around a keyword group, up from 9, and 58 keywords on Google page 1 instead of 6. Organic visitors grew from 640 to 1,710 a month, and 58% of them now landed on one of the leak problem pages, which did not exist at the start. Monthly organic enquiries rose from 7 to 23. The owner noted that enquiries from the problem pages usually arrived with photos of the leak attached, which made quoting faster.

MetricBaselineMonth 6Change
Keywords mapped to a page0386+386
Pages built around a keyword group923+156%
Keywords on Google page 1658+867%
Monthly organic visitors6401,710+167%
Share of organic visits landing on leak problem pages0%58%+58 pts
Monthly organic enquiries723+229%

Month 0 is the month before the engagement. Rankings from SEMrush position tracking, visitors and landing pages from Google Analytics, enquiries from the client's form and WhatsApp log.

Months 1 to 2
Research and mapping

212 customer messages read, 1,140 phrases collected, 386 mapped to pages in 4 intent groups.

Months 3 to 4
Page build

14 new pages published and 5 existing pages re-pointed. Enquiries reached 12 a month by Month 4.

Months 5 to 6
Tracking and gaps

11 keywords moved to better pages, 23 new phrases added. Enquiries reached 23 a month.

Key Takeaways

What We Learned

Read your enquiries before you open a keyword tool

If you run a trade or service business in Singapore, your WhatsApp and email history is the best keyword source you own. It shows how customers describe the problem in their own words, and those words are usually the searches you are missing.

Name pages after the problem, not the method

Customers do not search for the technique you use. They search for what is going wrong. A page called PU injection reaches other contractors. A page about a wall that leaks in heavy rain reaches the person who needs PU injection.

Who pays is a buying question

Questions about responsibility look like research, but they come from people with a live problem. Answering them clearly, with the next step at the end of the page, brought in some of the client's best jobs.

FAQ

Frequently Asked Questions

For a business with one core service and a few related ones, the research and the keyword map usually take three to five weeks. Here it took about six weeks, because we read the client's enquiry history as well as the search data. Building the pages from the map takes longer, and that depends on how many are needed.
You can use them as a starting point, but they only show what your competitors chose to target. In this case none of the competitor sites had pages about who pays for a leak, and that group ended up producing some of the best enquiries. Your own customer messages will show gaps that competitor data cannot.
Usually both. Existing pages that already rank for something should be re-pointed rather than replaced, because they carry history with Google. New pages are only worth building when a group of searches has no good home on the site. This client kept and re-pointed 5 pages and built 14 new ones.
Research and mapping is a small part of the total cost, and it decides where the rest of the budget goes. We quote it as a fixed project or as the first month of an SEO engagement, depending on how many services and locations you cover.

Want Results Like This for Your Service Business?

Get a free SEO audit. We will look at the searches your customers use, the pages you have now and the gaps between the two.