A marketing plan that fits on one page

Start with where your last fifty jobs came from

Open your invoices and write down the source of your last fifty customers. Referral, Google search, the map listing, paid ads, a truck somebody saw, a neighbor, the town Facebook group. You will not get all fifty right, and that is fine. The pattern is what you are after, not the decimal places.

Almost every owner who does this is surprised by something. Usually it is that referrals are carrying more of the business than they assumed, or that the channel they have been paying for every month is producing about three jobs a quarter. That list is worth more than any market research you could buy, and it costs an hour of your time.

The number that sets your budget

Percent-of-revenue budget rules are close to useless for a business with fifteen employees. The number you need is the most you can pay to get one new customer without losing money on them.

The arithmetic is short. Say an average job is 12,000 dollars at 35 percent gross margin. That is 4,200 dollars of gross profit. If you close one estimate in four, then four leads have to produce one job, so four leads have to cost meaningfully less than 4,200 dollars for the channel to be worth running. Divide it out and you have a ceiling per lead. Every channel gets judged against that one number, not against each other.

Run the same math for your worst service line and your best one. Most owners find one service they can afford to advertise aggressively and one they should never pay for a click on.

Pick one thing to fix and one thing to test

A plan with six channels is a wish list. Most small businesses can execute one improvement and one experiment per quarter, and that assumes the owner is involved in both.

The fix is usually a channel already producing work that is being handled badly: the Business Profile with four reviews, the website that never says which towns you cover, the ad account counting form submissions while ignoring phone calls. The test is the new thing, with a fixed budget and an end date. If the test has not produced a customer by the end of the quarter, it dies quietly and without a meeting.

What actually goes on the page

• The two or three services you want more of, in order. Not everything you are capable of doing.

• The towns you want work in, ranked by drive time and job value.

• Your maximum cost per new customer, from the arithmetic above.

• This quarter's one fix and one test, each with a person's name next to it.

• The three numbers you will check monthly: leads by source, close rate, average job value.

• What you are going to stop doing.

The last line is the one people skip, and it is the one that makes the plan real. A plan that only adds work is a list of things you will feel guilty about in April.

Where these plans usually fall apart

Month two, almost always, and for one of three reasons. Nobody owns the work, so it becomes whoever has a slow afternoon. The lead source question stops getting asked at booking, so by month three you are guessing again. Or somebody sells the owner something on a cold call and the budget quietly moves.

The fix for the first two is boring and it works: put the words how did you hear about us into the booking script as a required field, and put one person's name on each line of the plan. The fix for the third is the plan itself. It is much easier to say no to a directory listing salesperson when you can point at a page that says what you are doing this quarter.

Reviewing it without buying software

You do not need a dashboard. For a service business the honest source is the phone. Use a tracking number on the channels you pay for, ask the question at booking, and keep a twelve-row spreadsheet with the month, leads by source, jobs closed, and spend.

Analytics has its place for a site that sells online, and there is a real case for setting up conversion tracking properly if you are running paid search. But a spreadsheet somebody actually fills in beats a reporting tool nobody opens.

What the plan cannot do

It will not fix capacity. Generating leads you cannot answer within an hour is an expensive way to make people angry, and we have watched more than one campaign get blamed for a problem that was really a phone going to voicemail at 4:30.

It will not manufacture demand that is not there. If nobody in your area is searching for the service, no channel plan changes that, and the honest answer is to sell something else or go wider geographically.

And attribution stays messy. Expect a chunk of your customers to say they are not sure how they found you, or to name the last thing they touched rather than the first. Do not spend money trying to close that gap. Get the pattern, act on it, and revisit the page every quarter.