Local service marketing metrics: score each channel on five rows

What to take away

Why a written rubric settles the budget argument

A written rubric settles the budget argument before it starts. Every channel gets the same five questions each month, so the owner's favourite channel and the loudest voice in the room no longer win by default. The decision is where next month's money goes.

Test each proposed row first: ask what you would stop doing if the number halved. If the answer is nothing, the metric is decoration. A number that cannot change a spending choice is a hobby, and the rubric has room for only five rows.

The wider framework for building the measurement behind these rows is in measurement and reporting. The rubric sits on top of it and answers one narrow question: which channel gets the next pound. It assumes the enquiries are already logged in one place.

What sits beneath each row

Each row is fed by one number from the enquiry log, so the rubric is only as good as the log. A channel whose enquiries arrive untagged cannot score above the bottom row for traceability. Untagged counts are the kind of measurement mistakes that cost leads.

A channel that sends twenty enquiries and wins two is worth less than one that sends six and wins three. Booked-job rate exposes that difference, which is why raw enquiry volume does not get a row of its own.

For cost per job, GOV.UK's Service Manual works out cost per transaction as the total cost of providing a service through all channels, divided by the total number of completed transactions.

The same shape suits a booked job. The hours spent answering enquiries belong in the total, not only the ad spend, and local service marketing costs shows what those costs look like in England.

Score each channel out of five

Score each channel monthly out of five on each row, total the column, and compare channels rather than months unless your seasonality is strong. The descriptors below anchor a 5 and a 1, and the scores between them are your judgement.

Set your own thresholds. The rubric forces a conversation about evidence rather than opinion. Write the thresholds down before the first scoring session, so nobody adjusts them after seeing which channel they favour.

Tags show where an enquiry arrived, not why it came. How far to trust a tag is covered in choosing an attribution method. Borrowed averages need the same care, and how benchmark research is built explains why.

Run the scoring on the same date each month

Review monthly, on the same date, and compare each channel against its own previous three months. Flag any channel that drops for two months running, because a single weak month may be seasonality and two in a row calls for a decision.

Keep the report to one page. A page that gets read beats a deck that does not. The reporting dashboard shows how to lay the scores out so the monthly review takes minutes.

GOV.UK's guidance on using performance data says it is important to use the data collected to find ways to improve the service and prioritise them. For a small firm, that prioritising is the budget decision the rubric exists to support.

Common questions

How many metrics should a small firm track?

Four to six. Enquiry volume by source, response time, booked-job rate and cost per booked job cover most decisions. Add one channel-specific figure if you are testing something new, and drop any figure that supports no decision.

What should happen when two channels score the same?

The channel with the cheaper booked jobs takes the next round of budget. A tie in the rubric is a reason to look at cost per job, not a reason to split the money evenly and stop asking the question.

Should I compare months or channels?

Compare channels, not months, unless your seasonality is strong. To judge a single channel's trend, compare it with its own previous three months and flag any drop that lasts two months running.