Bid the build season backward: money moves in January for a June dig
Budget in this market should follow the excavation calendar, not spread evenly across twelve months.
Freeze risk and a shortened dig window mean the season has real ends. The contracts that fill a summer get signed in late winter and early spring, so the auction that matters most happens while the yard is still brown.
Front-load accordingly. Push hard from midwinter through late spring, hold a lighter presence over the summer for renovation and repair, and make a deliberate decision about whether to buy at all in the dead weeks rather than letting the account idle.
Pull spend down when the schedule fills. An account still buying new build clicks after the season is booked is paying for inquiries you will have to decline, and declined inquiries cost you goodwill and sometimes a review.
Renovation, equipment and cover work follow a different curve. Give them their own budget and their own on and off calendar so they are not competing with construction in the same campaign.
A twenty mile ring around Bethesda buys clicks in two states and a federal district
Radius targeting is the most expensive default setting in a DMV pool account.
Draw a circle around an office in Bethesda or Tysons and it crosses the District line, the Maryland line and the Virginia line. Each of those licenses trades separately, permits separately and taxes separately. Clicks from a jurisdiction you do not serve cost exactly the same as the ones you want.
Target by jurisdiction instead. Separate campaigns, or at minimum separate ad groups, for the District, for Montgomery County and Prince George's, and for Arlington, Alexandria and Fairfax County.
Splitting the geography also makes the copy honest. An ad that names Falls Church or Reston reads local to someone in Falls Church or Reston, and it filters out the searcher you cannot help before they cost you anything.
The split produces the only report worth having: performance per jurisdiction. Blended metro numbers hide the market that is subsidizing the one that is not working.
Pool opening, closing and weekly service searches will drain a construction budget
Service demand in this metro is steady, well funded and worth nothing to a firm that builds and renovates.
Opening and closing, weekly maintenance, filter and pump cleaning, chemicals, algae, leak checks on a pool you did not build. All of it converts, in the sense that people click and call. None of it fills a dig calendar.
Build the negative list before launch rather than after the first invoice, then read the search terms report weekly for the first two months. In a market this dense the junk arrives fast.
Add the categories a metro area generates on its own: municipal and community aquatic centers, apartment and condo amenity pools, swim lessons, lifeguard jobs, hotel pools and pool hall listings. Every one of them shares your vocabulary.
In a small account, negatives improve results faster than bid changes. Cutting waste raises the effective budget without asking the owner for another dollar.
Hot tub and swim spa shoppers look exactly like your buyers and are not
Real winters make spas a strong seasonal category here, which puts retail shoppers into a builder's auction.
Searches for hot tub installation, spa delivery and swim spas rise as the weather turns. The intent behind them is retail purchase, often with a dealer, not a gunite project with a permit and an inspection sequence.
If you build spas as part of your construction work, that deserves its own campaign, its own copy and its own price framing, so the message matches what you actually sell.
If you do not, exclude the vocabulary completely rather than hoping the landing page sorts it out. Every one of those clicks is paid for before anyone reads a word.
Apply the same reasoning to above ground pools, indoor pools and stock tank conversions. Deciding what you are not is half of a functioning keyword list.
Cost per signed design agreement, counted separately for each jurisdiction
A blended cost per lead across three jurisdictions hides which market is actually paying for itself.
Leads are easy to count and tell you very little. Track the milestone that means money: a signed design agreement, a paid deposit, a scheduled dig. Feed it back to the ad platform through offline conversion import so the bidding learns from outcomes rather than form fills.
Report it by jurisdiction. A job on a District lot with alley access, tight staging and a compact machine carries different costs and different margin than a Fairfax County subdivision with a driveway, and averaging them together tells you nothing you can act on.
The sales cycle is long enough that the month of the click and the month of the signature are frequently different months. Report on cohorts, tracking the clicks from a given period through to what they eventually signed, or the numbers will lie in both directions.
Buyers here often run procurement-style diligence. Expect requests for license verification, references and written scope before signing, and build that into your expected time to close rather than trying to short circuit it in an ad.
Pacing spend against one excavator, a permit queue and a calendar that ends at frost
Spend that outruns your crew produces voicemails you cannot return, not revenue.
Capacity in this trade is physical. One crew, one machine, subcontractors with their own books, and permit review running on three separate jurisdictional clocks.
Set a weekly ceiling on qualified inquiries you can genuinely visit and quote, and pace the daily budget to hold it. An account that beats that ceiling is manufacturing a backlog of disappointed people.
Winter weather removes working days without warning. Snow and ice response, frozen ground and delayed inspections all compress the schedule, so leave slack in both the calendar and the budget plan.
When the season books out, shift money rather than switching everything off. Renovation, equipment upgrades and next season deposits absorb spend productively and keep the account's learning intact.
Questions we actually get
- Should we run one account for the whole metro or split it by jurisdiction?
- Split it. The District, suburban Maryland and Northern Virginia are separate legal jurisdictions with their own permitting and licensing, and a radius that ignores that will buy clicks you cannot serve. Separate campaigns let you write geographically honest copy, set different budgets where margin differs, and produce a cost per signed job for each market instead of an average that hides the weak one.
- What negatives matter most for a pool construction account here?
- Service vocabulary first: opening, closing, weekly maintenance, chemicals, algae, leak checks. Then retail spa and hot tub terms, which get expensive as the weather turns. Then the metro's own noise: community and municipal aquatic centers, apartment amenity pools, swim lessons, lifeguard jobs, hotel pools. Build the list before launch and review the search terms report weekly for the first couple of months, because the junk arrives quickly in a dense market.
- When should we start spending for next summer's work?
- Earlier than most firms do. Because the dig window closes with the cold, the contracts that fill a summer are typically signed in late winter and early spring, which means the paid activity that matters happens while the ground is still frozen. A common structure is heavy spend from midwinter into late spring, lighter always on presence over the summer aimed at renovation and repair, and a deliberate decision about the dead weeks rather than a passive drift.
- Is cost per lead a fair way to judge the account?
- It is fair for diagnosing waste and misleading for judging value. A cheap lead from a jurisdiction you do not serve is worse than an expensive one that signs. Track through to the design agreement or deposit, import those conversions back into the platform, and report by cohort because the click and the signature often fall in different months. Report separately by jurisdiction as well, since costs and margins genuinely differ.
- How do we keep paid spend from outrunning the crew?
- Decide the number of qualified site visits you can actually run in a week, then pace budget to that number rather than to a monthly total. Weather in this market removes working days without notice, so leave slack. When the season books out, move the budget to renovation, equipment work or next season deposits instead of pausing everything, which preserves the account's performance history.
