The short version.
A vendor paid per lead, or paid a retainer to manage what is inside the ad account, reports the numbers the ad account can produce: leads, cost per lead, click-through, the platform's own lead count. Those numbers stop at the form. The question you are actually asking, "did the money I spent turn into appointments my team could close?", is answered by counts that sit in the calendar and the CRM, and the vendor's job description ends before it reaches them.
So both sides settle on cost per lead, because it is the only number the contract can defend and the platform can produce. You end up paying for performance that exists inside an ad account while the calendar the ads are supposed to fill stays at the same number.
That is not a bad vendor. That is the lead-vendor model working exactly as designed.
Why the report looks fine while the calendar does not.
Cost per lead is the most flattering number in the route, for a mechanical reason: it is spend divided by the largest count. Every stage after it makes the count smaller and the cost per unit larger. Of every hundred leads, some share books, a smaller share attends, a smaller share was ever a fit, and a small number buys. A report built on the first count can improve every month while the last count stays flat, and it will, because the cheapest leads to buy are the ones least likely to book.
The platform makes it worse in a way nobody chose. Meta optimises toward whatever event you tell it counts. On a lead campaign, that event is a form submission. Loosen the form and the platform gets very good at finding people who submit forms, which is not the same population as people who book an estimate and show up for it. The vendor's number improves. Yours does not. Why the platform's lead count and your calendar can never agree.
Why the next vendor inherits the same problem.
You already know the cycle. Month one: onboarding, a new account structure, a fresh set of ads. Months two and three: "the algorithm is learning." Months four to six: cost per lead is down, the calendar is not up. Month seven: you are on another sales call watching another case-study deck.
The next vendor inherits the exact same job description. Manage what is inside the ad account. Report the numbers the platform produces. Defend the fee with cost per lead, because that is the only number the contract can defend. The measurement problem, the follow-up problem and the show-rate problem all survive the handover untouched, and the new team optimises toward the same number the old team did, with a fresh ninety days of honeymoon to spend doing it.
Firing the vendor and hiring a better one changes the people and keeps the model. If you are asking whether yours is worth keeping at all, take the 9-signal test first; it separates a vendor problem from a measurement problem, which is the distinction that decides what to do next.
What a structural fix looks like.
Three things have to change, and none of them is "better ads."
- Success is defined on a number you can verify yourself. Attended appointments by source. Cost per attended appointment against what one is worth, which you derive from your own revenue per client, margin, close rate and show rate. The Allowable Cost per Appointment Calculator draws that line in a minute.
- One party owns the route after the click. The page, the application, the follow-up in the first hour, the booking path, the confirmation and reminders, the qualification, the handoff to your closer, and the measurement across all of it. Where the route is owned in pieces, it breaks at the seams nobody owns, and the piece that reports first wins the argument.
- The report reconciles to the calendar and the CRM. Platform leads, CRM contacts, bookings, attendance, qualification and closes, by source, in one table, with the gaps explained. Revenue comes from the payment processor, never from the CRM.
A vendor who does those three things is not a lead vendor any more. They are operating your appointment engine, and they should be measured like one.
What Tran Growth Partners built instead.
Tran Growth Partners builds and operates the whole Meta-to-appointment route as one system: the offer diagnosis, the creative, the campaigns, the page, the application, the follow-up, the booking, the reminders, the qualification logic and the measurement across all of it, reported on the counts in your calendar and CRM. Your team owns the sales conversation and the close. The front-end commitment is thirty qualified appointments with your ideal prospects who actually show up within ninety days, or Tran Growth Partners keeps working for another ninety days without charging its service fee. The waiver covers the service fee only, and the exact definitions are agreed in writing before launch.
Start with the free version of the discipline. The Appointment Engine Scorecard is the five-number check Tran Growth Partners runs before any budget decision: cost per attended appointment against the allowable, show rate, qualified rate, booking lag and calendar utilisation, with a scale, hold, fix or pause verdict.