The Recon Method: we know your market before we spend your first dollar.

Before we look at a single competitor, we pin the numbers everything else is judged against: your average job value, your close rate, your lead-volume goal, your legal geography — and most importantly, what you pay for leads today.

We compute your incumbent bar: the all-in cost per lead of your current channel — a lead broker, a home-services marketplace, a referral network — and whether those leads are shared or exclusive. That single number becomes the pass/fail line for every channel we research. An exclusive $100 Google lead decisively beats a shared $95 marketplace lead, and your plan says so in your numbers.

Your memory of your past marketing is unreliable — so we verify it the same way we verify competitors: your complete ad history from Google's Ads Transparency Center and Meta's Ad Library, your live tag configuration, and your old landing pages from web archives.

For a recent outdoor-living client in the Dallas–Fort Worth market, this phase found exactly one Google ad ever run — dead for a year, pointed at the homepage, with zero conversion tracking behind it. The phase ends with one sentence that reframes everything: “‘Google Ads didn’t work for us’ is fully explained by the absence of measurement, not the market.”

Three sweeps, because each finds companies the others miss: a paid sweep of your top commercial-intent searches, a local and organic sweep of the map pack and page one, and your own list of who you actually lose bids to.

Every name lands in one of three buckets — active advertisers (who you'll bid against), organic incumbents (the trust bar you must match), and sleepers (real competitors invisible online, whose market share goes to whoever shows up first). We also run a name-collision check: a near-identical competitor name changes how your brand campaign gets built.

For every active advertiser, we pull their actual ads from the platforms' own libraries — never inferring from their website what they “probably” run. Ad counts with pull dates, headlines quoted verbatim, the offers and warranties they push, and whether their ads land on dedicated pages or homepage dumps.

Then we mine the gaps. What everyone says (“free quotes”) becomes table stakes we include but never lead with. What no one says becomes your headline lane. Competitor ads and search results also reveal the junk intent to exclude — the DIY, jobs, and bargain-hunter searches that quietly burn most contractor ad budgets. Your negative-keyword lists are written here, before launch, not discovered after three months of wasted spend.

Published industry cost data — labeled as the national planning ranges they are — goes up against your incumbent bar and a simple affordability check: average job value × close rate × the marketing percentage you can tolerate.

The result is a staged budget, not a number pulled from the air: Prove (minimum viable spend on your top two or three services — calibration, not verdict), Scale (reallocate to measured winners, draw down the incumbent channel only where we beat its bar for 30 straight days), Dominate (brand defense, broader terms, remarketing — only after real creative assets and proven cost-per-lead exist).

Research that stays in someone's head is worthless. You get a numbered, sixteen-section document you can read and we can build from: your economics, your verified ad history, the competitor dossier, keyword and negative lists, written ad copy, landing-page mapping, measurement spec, and a launch checklist where every open item has a named owner.

Three rules make it trustworthy. Message match: the pinned ad headline mirrors the landing page it points to, campaign by campaign. Claims discipline: no offer, warranty, review count, or city appears in copy until it's confirmed in writing. And the launch gate: nothing spends until one end-to-end test passes on a real phone — click, tracked call, and the lead visible in Google Ads, analytics, and your inbox. Recorded.

A monthly transparency-center re-pull watches rivals enter, exit, and change message. Weekly search-terms reviews for the first 60 days feed the negative lists. Once campaigns run, auction insights replace outside-in research with the platform's own overlap data.

And at day 30, your measured cost-per-lead replaces every benchmark number in the plan. From then on, decisions run on your data, not the industry's.