Thirty-three leads came in last week. Fifteen booked an appointment. Seven actually showed up.
That was one metro market for a home services client of ours, read off their own scorecard on a call this week. The owner's line stuck with me. "We're cutting it in half twice before we ever get to our close rate."
His sales reps had an explanation ready. The leads are bad. People are calling about recessed can lights when the company sells permanent outdoor lighting. The office had a different explanation. The reps are slow and the follow-up is inconsistent. Two agencies were on the call, one running Google and one running Meta, and both of us were quietly wondering whether it was our fault.
A low lead to appointment conversion rate is one of the most common problems I see in lead generation businesses, and it is almost always misdiagnosed. This post is the process I actually use to figure out whether the problem is the leads or the follow-up, because the fix is completely different depending on the answer.
What a lead to appointment conversion rate actually measures
Most businesses track one number here and call it good. That is the first mistake. There are three separate rates between a lead and a sale, and they fail for different reasons.
Set rate. Of the leads that came in, how many booked an appointment? This is the number people usually mean when they say lead to appointment conversion rate.
Show rate. Of the appointments that were booked, how many actually happened? Some teams call this keep rate.
Close rate. Of the appointments that happened, how many bought?
In that example that is a 45 percent set rate, a 47 percent show rate, and a close rate hovering around 30 percent. The owner wanted a set rate of 60 to 75 percent and a show rate near 100. His words were that anything under 40 percent on set rate is "complete garbage," and anything around 50 is hard to read either way.
For context, Invoca's 2025 benchmarks report, built on more than 60 million calls, puts home services call conversion at 46 percent and found that only 35 percent of agents actually ask the lead to book or buy. Those two numbers together tell you a lot. Roughly half of leads converting on the phone is normal, and a big chunk of the leak is the person answering never asking for the appointment.
The reason to split the three rates apart is simple. Lead quality problems show up in set rate. Operations problems show up in show rate. When you blend them into one number, you lose the ability to tell which one you have.
Why both sides are usually half right
I have sat in a lot of rooms where sales says the leads are garbage and marketing says sales never works them. In my experience both are partly true almost every time, and arguing about which is more true wastes weeks.
Bad leads and bad follow-up compound each other. A rep who has been burned by three unqualified calls in a row starts calling the fourth one slower and with less energy. A great lead that waits a day for a callback books somewhere else, and now they look like a bad lead in the CRM.
So the question is never "is it the leads or the follow-up." The question is which one is the bigger leak right now, and what evidence would settle it. Here is the order I work in.
Step one, make sure the number is real
Before you diagnose anything, confirm the rate you are looking at is accurate. This sounds obvious. It is skipped constantly.
On that same call, the client had just rolled out a new dashboard. It pulled appointments from the sales pipeline, so every time an opportunity was created it counted as an appointment set. The office KPI sheet counted appointments from the actual call outcome logged by the rep. The two disagreed.
When we clicked into the dashboard's list, the same names were in there two and three times. Every time a rep updated a note or a lead called back the same day, a new pipeline entry was created. The "gross set rate" the owner had been staring at for three weeks was inflated by duplicates, which meant the true set rate was even lower than the scary number on screen. Nobody could trust any data since the dashboard launched.
Same week, a different client in financial services told me their ad platform was reporting about a third more booked meetings than they actually had. Fifty in the ads manager was really 33 to 35 in the calendar. The conversion event was firing on more than one button.
Both clients were about to make decisions off numbers that were wrong. One would have overestimated the problem, the other would have underestimated it.
Before you go further, do three things. Pull the raw list behind the rate and look for duplicates. Confirm what "appointment set" actually means in your system, whether it is an outcome a human logged or a stage a record landed in. And reconcile the ad platform's count against the calendar for one month. If the numbers do not match within a few percent, fix that first. I wrote about the mechanics of this in Why Your CRM Data Is the Most Valuable Asset in Your Ad Account.
Step two, split the leak into set rate and show rate
Once the numbers are clean, look at where the drop happens.
A low set rate with a healthy show rate points toward the front of the funnel. Either the wrong people are coming in, or they are coming in and nobody reaches them in time. Both of those are diagnosable, and I will get to how.
A healthy set rate with a low show rate is almost never a lead quality problem. These people called you, talked to a human, and agreed to a time. Then half of them did not show. That is an operations issue. The usual culprits are no reminder sequence, too many days between booking and appointment, no confirmation touch the day before, and appointments booked without the decision maker present.
On that call, the owner's own read was that a 50 percent show rate could not be a lead problem. "Why would half the people that book not keep? They called us." He is right. When someone books and does not show, the marketing already did its job. The system after the booking did not.
Set rate is where it gets more interesting, because it can be either. Which brings us to the tiebreaker.
Step three, listen to the calls
Call recordings settle this argument faster than any dashboard. If your office runs calls through a phone system or CRM that records, you already have the evidence. Most teams just never review it.
We are pulling August's recordings for that home services client right now, split by which channel the lead came from so each agency can see its own trends. Here is what I listen for.
Wrong product intent. The caller wants something you do not sell, or a cheaper version of what you sell. The can lights complaint is a real example. If this shows up a lot, it is a targeting and messaging problem and the fix lives in the ad account.
Competitor confusion. The caller thinks they are calling someone else, or comparison shopping off a search that matched a competitor's name. Also a targeting fix.
Price shock. The caller is qualified but was not prepared for the price. This is a landing page and creative problem. The page should be setting expectations before the call, not after.
No urgency. Qualified, interested, but "let me get HOA approval" or "bad timing." On that call a rep literally told the owner a customer wanted to wait for a full moon to pass. That is a follow-up cadence problem, not a lead problem, as long as the sequence keeps working the lead over the next 30 to 60 days.
The rep never asked. The caller was ready and the conversation ended without a booking attempt. Invoca's data says this is the norm, not the exception. Pure operations.
You do not need to listen to every call. Pull 20 to 30 per channel from a period where your numbers were trustworthy. If your phone system transcribes calls, run the transcripts through an AI tool and ask it to categorize each one into the buckets above. Then read a sample yourself to make sure it is categorizing correctly. Two hours of this beats a month of debate.
What the answer changes
If the calls say lead quality, the work moves to the ad account and the page. Tighten search terms and add the negatives you just learned about. Rewrite creative to qualify harder, even if it costs you volume. Add a question or two to the form that filters the wrong product intent before the call. Above all, feed the outcome back to the platform. If Google and Meta only ever hear "lead," they will keep finding you cheap leads. If they hear "appointment set" and "closed," they start finding people who book and buy. That feedback loop is the core of our framework, and it is also why the best conversion optimization sometimes adds friction on purpose.
If the calls say follow-up, the work moves to the office. Speed to first contact, ideally minutes. A written multi-touch sequence across call, text, and email that runs for weeks, not days. Reminders and a day-before confirmation for every appointment. A booking script that actually asks for the appointment. And a clear owner for each lead so nothing sits.
For the financial services client, the answer was a bit of both. Their lead to booked meeting rate sits around 8 to 9 percent and they want 15. Part of that is fixing the tracking so the platform optimizes toward real meetings instead of a button that fires whenever. Part of it is monthly conversion work on the booking pages themselves. We are doing both, and we will remeasure each month before touching the next thing.
The numbers to watch every week
Keep this on one page and review it weekly with both the marketing side and the office side in the room.
- Leads by channel, deduplicated
- Set rate by channel
- Show rate by channel
- Median time to first contact
- Close rate from appointments that happened
- Cost per appointment that showed, not cost per lead
That last one is the number I would run a business on. Cost per lead rewards volume and hides every leak downstream. Cost per kept appointment forces the ad account, the page, and the office to be judged as one system, which is what they are.
Yes, this is the whole playbook
There is nothing held back here. Clean the data, split set rate from show rate, listen to the calls, then fix whichever side the evidence points to. A business that runs this on its own and fixes its funnel was never going to hire us, and I am glad they read it.
The businesses that do hire us usually know all of this and do not have the time or the systems to run it every month, connect the outcomes back to the ad platforms, and keep the whole thing honest. That is what we do. If you want a second set of eyes on where your own leads are leaking, apply for a free system review and we will walk through your numbers together.
Sources
- Invoca Call Conversion Industry Benchmarks Report, 2025, analysis of 60 million calls, including the 46 percent home services conversion rate and the finding that 35 percent of agents ask leads to book or purchase.
- Supply House Times coverage of the Invoca report, June 2025.
About the author
Kyle Rutledge
Owner
I’m Kyle, founder of Gradari, a paid ads lead generation agency that helps B2B and SaaS companies stop wasting budget on low-quality leads and start building systems that actually drive growth.