Two pipelines, two clocks: a homeowner decides in a week, a board decides over a quarter
Applying one follow up rhythm to both buyers guarantees you are too slow for one of them and give up too early on the other.
Residential inquiries are a speed contest. Whoever gets a human on the line and a date in the calendar first is usually in the final two. Everything else about your marketing matters less than that.
Association inquiries are a patience contest. The manager may like you in week one and still be unable to act until a board meets, a reserve is confirmed or an engineer's report lands. A standard sequence that stops after three unanswered emails drops the job precisely when it was normal for nothing to be happening.
Build two cadences and label every opportunity with which one it belongs to at the moment of arrival. Residential: minutes to first contact, then a visit. Commercial: a slower, useful drumbeat timed around meeting dates, with a document to forward each time.
Ask managers directly when their board meets and when budgets are set, then put those dates in your own calendar. Following up the week before a meeting is worth more than five follow ups at random.
Where amenity deck inquiries leak: the site walk nobody scheduled
Commercial pool work rarely dies from a lost argument, it dies from a step that was never booked.
The most common failure is a proposal sent as an attachment with no scheduled next contact. The manager forward it, the board postpones, three weeks pass, and by the time anyone looks again there are two other quotes on the table.
The second is access. A site walk gets scheduled without arranging building access, the estimator arrives, the freight elevator is booked and nobody left a name at the desk. That wastes a morning and signals exactly the wrong thing about how you run jobs.
The third is a proposal missing what a board actually asks about: staging around residents, working hours, resident notice, protection of finished areas, insurance, and how the amenity level is returned to use. A beautiful scope that ignores those questions gets sent back for clarification, and clarification takes a month.
Fix all three with a rule rather than a reminder: no proposal leaves the office without a named owner, a scheduled next date and the resident impact section already written.
The first sixty seconds decide whether a Spanish-speaking caller stays on the line
Intake in a county where Spanish is the dominant home language across large parts of the market is an operational requirement, not a courtesy.
A caller who reaches an English only greeting will frequently hang up without leaving a message, and you will never see that inquiry in any report. It looks like traffic that did not convert.
Answer live in both languages during business hours. Record the voicemail greeting in both. Route Spanish form submissions to a bilingual responder, and make sure the automated confirmation email is in the same language as the form.
Follow through to the site walk. Sending an estimator who cannot discuss finishes and staging with the household undoes the good work done by everything upstream.
Listen to recordings monthly. The gap between how you think calls are handled and how they are actually handled is usually the single largest conversion finding in the business, and it costs nothing but attention to find.
What intake must capture before anyone drives to Doral
An estimator's day is the most expensive thing in the business, and a two question intake spends it badly.
Residential intake needs enough to qualify without interrogating: property type, the job as they describe it, rough timeline, a budget band and whether they own the property. Ask the budget question plainly and early; people answer it more readily than the trade expects.
Commercial intake needs a different set: building, association involvement, deck level, access, decision maker, and whether an engineer's report already exists. Getting the access question answered on the phone prevents the wasted morning described above.
Write the script down, keep it to one page, and make sure whoever answers the phone at four on a Friday has it in front of them. Consistency at intake is where most of the available gain sits.
Set a floor and hold it. If a job below a certain size is not worth a site visit, decide that once and route those inquiries to a referral partner or a phone estimate rather than deciding case by case in the moment.
Counting site walks and signed agreements instead of form fills
If your reporting stops at the inquiry, every downstream leak is invisible and every channel looks equally good.
Instrument the whole path: inquiry received, qualified, site walk booked, site walk completed, proposal issued, agreement signed. Record source and pipeline at the first step and carry both through to the last.
Use call tracking with dynamic numbers so phone inquiries land in the same ledger as forms. In this trade the phone is the primary channel and it is the one most often missing from the numbers.
Keep the two pipelines separate in every report. Blending a backyard resurface with an amenity deck job produces an average that describes no job you have ever done and hides which half is actually paying.
Watch the stage to stage ratios rather than the totals. The ratio that drops fastest is where the money is, and it is usually somewhere between the proposal and the signature rather than at the top.
With one metro's volume, change the script before you change the button
A single builder in a single county rarely produces enough conversions to detect the small effects that split testing was invented to measure.
Reserve testing for changes big enough to show up: replacing a long form with a short one plus a scheduler, offering a booked appointment instead of a callback, or adding a Spanish path where none existed. Small cosmetic tests will not resolve at your volume, whatever a dashboard claims.
Read the results over quarters, not weeks, and treat sequential comparisons as directional. Miami competitive conditions and your own advertising shift underneath any test that runs long enough to matter, so hold conclusions loosely.
Substitute qualitative evidence where statistics cannot help. Call recordings, session replays and one honest question to prospects who chose someone else will tell you more in a fortnight than a button color test will in six months.
Then fix the operational things that need no test at all: answering the phone, replying the same hour, booking the visit on first contact, and following up the proposal. Nobody needs a significance calculation to justify those.
Questions we actually get
- Our inquiry numbers look fine but we are not signing enough work. Where do we start?
- Start by measuring the stages you currently do not: site walks booked, site walks completed, proposals issued, agreements signed, split between residential and building work. The drop is almost never at the top. In a mixed book it is usually a proposal sitting with a manager who has no scheduled next conversation with you.
- How fast do we really need to respond to a new inquiry?
- For residential, as close to immediately as your staffing allows, because the household is contacting several builders in one sitting. For commercial, same day is generally fine, and what matters more is a scheduled next step and follow up timed to their board or budget calendar rather than to your own sequence.
- Is it worth paying for a bilingual answering service?
- In Miami-Dade, usually yes, if nobody in the office can reliably take a call in Spanish. A caller who reaches an English only greeting frequently hangs up without leaving a message, so the loss never appears in your reporting. Just make sure the service can capture the qualifying questions and pass them on properly.
- Should we A/B test our website?
- Only for changes large enough to detect at one builder's volume in one metro, such as replacing a form with a booking flow or adding a Spanish path. Small cosmetic tests will not reach a trustworthy result. Call recordings, session replays and asking lost prospects why they went elsewhere are better use of the same effort.
- How much of the improvement can you actually promise?
- None of it as a number, and be wary of anyone who quotes you one. We can commit to instrumenting the path from inquiry to signature, separating the two pipelines, fixing the intake and follow up gaps we find, and reporting honestly on what moved. The gains in this work are usually real and unevenly distributed, but they are not predictable in advance.
