How We Got The Pool Police Ranking on Google in 60 Days (Real Case Study)

An honest SEO case study for a small business

Most "SEO case studies" you read online are written by agencies selling $3,000/month retainers, and the numbers always look suspiciously round. Traffic up 10x. Leads up 5x. Revenue up, trust us.

This is not that. This is a real, ongoing engagement with a real pool cleaning service in Lakeway, Texas, using the same $500/month content engine we sell to every other client. Some of the numbers below are measured. Some are projected based on the trajectory we're seeing. I'll label which is which so you can decide for yourself whether the playbook is worth copying.

If you're a small business owner wondering whether local SEO still works in 2026 — when half the internet is AI slop and the other half is Reddit — this case study is for you.

The client: The Pool Police

The Pool Police is a pool cleaning and equipment repair company serving the Lakeway, Texas area, just west of Austin. The owner, Keith Mallette, has been running the business for 25+ years. He's the guy who actually shows up at your house, diagnoses a failing pump by listening to it, and has opinions about salt chlorinators that he'll share for free.

Keith is great at pool service. Keith is terrible at marketing. That's not an insult — it's why we exist.

When we took on the project, here's what The Pool Police had:

He was losing new construction neighborhoods to marketing-first competitors who couldn't tell a DE filter from a cartridge filter but had a nicer website.

Starting point (week 0): the honest metrics

Before we touched anything, here's what we measured:

  • "pool service Lakeway" ranking: position #47 (page 5 of Google)
  • Google Maps local pack: not present for any neighborhood search
  • Organic traffic: roughly 50 visits/month (branded searches, mostly people who already knew the name)
  • Leads from the website: 0-2 per month, mostly form spam
  • Indexed pages on Google: 5 (home, about, contact, services, a gallery)
  • Google reviews: 39 at 4.9 stars (genuinely good — this is the moat)
  • PageSpeed: 34/100 on mobile (ouch)
  • The reviews are the critical detail. Keith had earned a real reputation over 25 years. What he was missing was not reputation — it was distribution. Google didn't know who he was.

    Week 1-2: Foundation

    The first thing we built was a new website from scratch. Not a "redesign" — a full rebuild on Astro + Cloudflare Pages, which is what we use for every small business client because it's free to host and absurdly fast.

    Here's what went into the foundation phase:

    The new site

  • Astro + Cloudflare Pages, 100/100 PageSpeed on mobile and desktop
  • 67 pages total at launch (we'll break this down in a second)
  • Schema markup: LocalBusiness with 18 service areas, Review, AggregateRating pulling the real 39 reviews, and Service schema for each of 8 core services
  • Security hardening: HSTS, a strict CSP, OWASP Top 10 pass — boring but it's a small ranking factor and it keeps the site from getting hacked, which matters more for a small business than anyone admits
  • Email authentication: SPF, DKIM, DMARC set up so quote emails actually land in customer inboxes instead of spam
  • The 67 pages

    This is where most small business sites fail. They have one home page, one services page, and a contact form. Google has nothing to rank for neighborhood-level searches.

    We built a matrix:

  • 8 service area pages: Lakeway, Steiner Ranch, Cedar Park, Bee Cave, West Austin, Four Points, Hudson Bend, River Place
  • 6 core services: pool cleaning, equipment repair, green pool recovery, leak detection, pool inspection, pool school
  • Core pages: home, about, services hub, blog, contact
  • Every neighborhood page mentions real landmarks, talks about local water quality (Lake Travis affects source water hardness), and links to the specific services that matter in that area. Steiner Ranch has more new construction, so that page emphasizes start-up service. Hudson Bend has more waterfront property, so that page emphasizes leak detection near lake fluctuations.

    This is not a template swap. This is the thing competitors won't do because it's tedious. But tediousness is the moat now that AI can spin up generic content infinitely.

    Week 3-4: Content engine goes live

    Once the foundation was in place, we turned on the content engine. The goal: 2-3 articles per week, every week, forever. No sprints. No big pushes. Just a steady metronome.

    The topics weren't random. Every article came from SERP analysis: we looked at the top 5 results for a target keyword, figured out what they covered and what they missed, then wrote something more useful. Examples from the first month:

    The cedar pollen article is a great example of hyperlocal content. Nobody in Ohio cares. Everyone in the Texas Hill Country cares for about six weeks every winter when the mountain cedar dumps yellow pollen all over every pool in Travis County. That article ranks because it answers a real question that only matters in one geography — which is exactly the kind of content that both Google and ChatGPT love.

    Week 5-6: Google Business Profile optimization

    Here's the part where a lot of SEO agencies ghost. Google Business Profile is free, high-leverage, and tedious to maintain. Nobody wants to do it. We did it.

    None of this is technically hard. It's just work that doesn't happen on its own.

    Week 7-8: Rankings start appearing

    Around the 6-week mark, things started moving. This is earlier than typical — and I think it's mostly because of the 39 existing Google reviews. Google already trusted this business. What we gave it was a site worth ranking.

    What we measured between weeks 6 and 8:

  • "pool service Lakeway" moved from #47 to page 2, then to #8 on page 1
  • Google Maps local pack appearances for 5 neighborhoods: Lakeway, Bee Cave, Steiner Ranch, Hudson Bend, and Four Points
  • Organic traffic climbed from ~50/month to roughly 400/month (an 8x lift, most of it from long-tail neighborhood queries)
  • Indexed pages went from 5 to 62 (of 67 submitted — the rest will get picked up)
  • PageSpeed held at 100/100
  • Week 9-12: Compounding (projected trajectory)

    This is the honest part. We're at roughly 60 days in as of writing, and the trajectory is clear but not complete. Here's what we're seeing and what we expect to see over the next 30 days based on how compounding normally works with this playbook:

  • "Emergency pool cleaning [neighborhood]" long-tail queries are starting to rank — these are high-intent searches that convert directly to calls
  • ChatGPT and Perplexity have begun surfacing The Pool Police for Lakeway-area pool questions (we tested 20 prompts; it showed up in 6, up from 0 at week 0)
  • Organic leads: trending from 0-2/month toward an expected 15-20/month as the neighborhood pages age and Google gets more confident
  • The rule of thumb on content-driven local SEO is that the first 60 days are foundation, the next 60 days are the real harvest. We're at the transition point.

    Key metrics at 60 days

    Pulling it all together, here's where The Pool Police is right now:

  • Google Maps local pack: present in 8 neighborhood searches (up from 0)
  • Organic traffic: approximately 8x the starting baseline
  • Ranking pages: 67 submitted, 62 indexed (up from 5)
  • Leads: tracking toward a 10-15x increase from the ~0-2/month starting point
  • Total cost: $500/month × 2 months = $1,000. The new website was included at no additional charge.
  • Compare that to what Keith was quoted by other agencies before he found us: one wanted $4,500 to build a new WordPress site and $2,500/month ongoing. Another pitched HomeAdvisor and Angi leads at $40-80 per lead, with no compounding benefit when you stopped paying.

    What actually made it work

    Looking back at the engagement, here's what I'd credit:

  • Neighborhood-specific pages, not just a city page. Steiner Ranch is not Lakeway is not Hudson Bend. Google knows this. Customers know this. Most competitors don't.
  • Consistent content publishing on a schedule. Not 20 articles in month one and then silence. A steady 2-3 per week, forever.
  • Schema markup done right. Real LocalBusiness, real AggregateRating pulling real reviews, real Service schema. Not the half-broken JSON-LD most small business sites ship with.
  • A fast site. 100/100 PageSpeed isn't a vanity metric — it's the difference between Google crawling 67 pages in two days versus two months. You can test your own site with our free website speed test.
  • Real reviews as the trust anchor. Keith already had 39 of these. If he'd had zero, the timeline would have been closer to 120 days than 60. This is the part you can't buy.
  • What didn't work (because nothing is clean)

    Three things I wish we'd done differently:

  • The first content calendar was too generic. We started with topics like "pool maintenance tips" that nobody searches for specifically. We had to pivot to hyperlocal angles (cedar pollen, Lake Travis water levels, hill country heat cycles) before traffic really moved. Lost about two weeks to that.
  • The initial meta descriptions were weak. We let the AI write them without a clear CTA style guide. Click-through rate on the first batch of indexed pages was mediocre. We rewrote them in week 5 and CTR jumped.
  • We didn't set up proper conversion tracking until week 3. Which means we probably missed counting a handful of early leads. Set up your tracking before launch, not after. Learn from us.
  • Could your business do this?

    Yes, probably — if you check three boxes.

  • You're a real business with a real reputation. At least a few years in, some existing reviews, an actual service you can deliver well. This playbook amplifies reputation; it doesn't manufacture it.
  • You're willing to wait 60 days. If you need leads by next Friday, buy Google Ads. If you want a compounding asset, do SEO.
  • You're serving a specific geography or niche. Hyperlocal content is the moat. If you're targeting "all of Texas" or "the whole internet," this playbook needs different math.
  • You can also do most of this yourself if you have the time. We're not gatekeeping. Start with our free schema markup validator to check your existing structured data, then run our free website security scanner to catch any technical issues holding your rankings back.

    Ready to run this playbook for your business?

    If you'd rather not spend your nights learning schema.org and SERP analysis, that's exactly why we exist. The Pool Police engagement is running on the same $500/month content engine we offer every small business client — with the new website built in at no extra cost. No contracts, no setup fees, no $2,500/month retainer.

    Start your own case study. We'll look at your existing site and tell you honestly whether this playbook will work for you — before you pay us anything.

    FAQ

    How long does local SEO actually take to work for a small business?

    In our experience, 60-90 days for initial ranking movement and Google Maps local pack appearances, and 4-6 months for the full compounding effect. The Pool Police moved faster than typical because they had 25 years of reputation and 39 real reviews as a trust anchor. A brand-new business with zero reviews should expect 90-120 days to see comparable traction.

    Why build 67 pages instead of just 5 or 10?

    Because Google can only rank pages it has. A single "service areas" page with 8 neighborhoods listed in a paragraph will not rank for "pool service Steiner Ranch" — there's nothing for Google to match against. A dedicated Steiner Ranch page that talks about Steiner Ranch specifically will. Every neighborhood you serve deserves its own page, and every service you offer inside that neighborhood is another valid page.

    Is $500/month really enough for local SEO in 2026?

    For a local service business with an existing reputation, yes. For a national e-commerce brand fighting Amazon for generic product keywords, no. The $500/month works because AI handles the heavy lifting on keyword research, SERP analysis, and content production, leaving humans to do strategy and quality control. We broke down the full math in our honest $500/month SEO breakdown.

    Do I need 39 Google reviews before this will work?

    No, but you need some. Five real reviews is a workable starting point. Zero reviews is a harder climb — you'll need to generate a few before Google treats the business as legitimate enough to rank locally. If you're at zero, start asking every happy customer for a review before you start a content engine. It's the cheapest SEO work you'll ever do.

    What happens if I stop paying after 60 days?

    Your existing rankings don't vanish overnight, but they stop compounding. The 67 pages we built stay indexed. The schema markup stays valid. Rankings typically plateau and then slowly erode over 6-12 months as competitors publish new content and you don't. This is different from paid ads, which stop generating leads the day you stop paying. SEO decays slowly; ads stop immediately. That's the trade-off.

    VK
    Vladimir Kamenev
    Founder

    25 years in industry

    Want us to build your website free?

    Custom website + 30+ SEO articles/month + AI search optimization. $500/month, no contracts.

    Get Your Free Website →