Every piece of software sold to a service business promises visibility. Dashboards, funnels, attribution models, colour-coded charts of things you did not know you could count.
You do not need any of it. What you need is five columns and twenty minutes a month, and the version on paper is frequently more useful because you have to look at every row.
The five columns
One line per enquiry, recorded when it arrives rather than reconstructed later.
Date. When they got in touch.
Source. Where they came from, in your own words. Google, Facebook group, yard sign, referral from Dave, the sign on the truck. This is the intake question, asked on every call and every form.
Spoke to them. Yes or no. Did an actual conversation happen, or did it end at a voicemail.
Quoted. Yes or no, and the amount.
Won. Yes, no, or still open, and the final value if it closed.
That is the whole system. A notebook works. A spreadsheet works better because it sorts, but the notebook version has been running businesses for a century and it beats software you do not open.
The four numbers that come out of it
At the end of the month, count.
How many enquiries. How many became conversations. How many became quotes. How many became jobs.
Four numbers, three gaps between them, and the gap that surprises you is the thing to fix. That is the entire diagnostic, and it is more actionable than any dashboard because it points at one specific problem rather than displaying forty metrics.
Most owners doing this for the first time find one of two things. Either the drop between enquiries and conversations is large, which is the missed call problem and is fixable this week. Or the drop between quotes and jobs is large, which is the follow-up problem and is fixable this week as well.
If the ratios are healthy all the way down and the top number is simply small, that is different information entirely. That is when a marketing budget is the right response rather than a distraction.
Add source and the whole thing gets sharper
Once you have a month or two, sort by source and run the same counts per channel.
You will almost certainly find that one source produces enquiries that rarely convert, and another produces fewer enquiries that convert reliably. Those two channels are not comparable on volume, and every dashboard that reports leads rather than jobs will rank them wrongly.
Then add what each channel cost you that month, including your own hours, and you have cost per booked job by channel. That single figure is what marketing software is ultimately trying to produce, and you have arrived at it with a spreadsheet and an intake question.
Use ninety days, not thirty
One month is too noisy to read. A single large job distorts the value column, a holiday week distorts the volume, and channels report at wildly different speeds.
Paid search shows up immediately. Content and local SEO take three to six months. A yard sign works whenever somebody happens to walk past. If you judge everything on a thirty day window, the slow channels always look like failures right up until they do not.
So keep the record monthly and read it quarterly. Rolling ninety days smooths most of the noise without hiding a genuine trend.
Compare against the same month last year
This is the correction almost everybody skips, and it matters more in the trades than in most businesses.
Home service demand swings enormously by season, with peak-to-valley variation of 250 to 600 percent in some trades and 40 to 60 percent month to month being normal. Comparing February to January tells you about winter, not about your business.
Once you have twelve months of records, compare each month to the same month a year earlier. That is the only comparison that separates a real change from the weather.
Which is also the argument for starting now rather than when you have time. The record is worth nothing in month one and a great deal in month thirteen, and there is no way to shortcut the wait.
Watch out for measuring what is easy
The natural failure is tracking the numbers that are simple to obtain rather than the ones that matter.
Website visits are easy and mostly meaningless. Follower counts are easy and mostly meaningless. Impressions are easy, meaningless, and reported enthusiastically by anyone selling advertising.
Jobs, job value, and where each one came from are harder to capture and are the only three that connect to the bank account. If a metric cannot change a decision you would actually make, it is decoration.
The monthly twenty minutes
Sit down once a month, ideally the same day each time, and do four things.
Count the four numbers and note them somewhere permanent. Look at the biggest gap and ask what caused it. Sort by source and check whether anything is producing enquiries that never close. And look at any channel you are paying for and ask whether the jobs it produced covered the spend.
Then write one sentence about what you will change. Not five. One.
Why this beats the software
Not because the software is bad. Because the record is only useful if it is complete, and a system that requires you to maintain it in an app you open twice a month will be incomplete within a quarter.
The intake question and the five columns survive busy weeks. And having to write the source down every time forces you to actually ask it, which is the single most valuable habit in this entire exercise.
Everything sophisticated in marketing measurement is an attempt to recover information that was never captured. Capture it at the source and the sophistication becomes unnecessary.