The symptom
You boosted a post because a slow week made you nervous. You are on a directory listing you cannot remember signing up for. Someone rang about SEO and you said yes to a trial. There is a Facebook page nobody has posted on since spring.
None of it is wrong exactly. It just is not connected to anything, so you cannot tell what worked, which means next month's decisions get made the same way — by whichever salesperson called most recently or whichever week felt quietest.
The fix is not a bigger budget or a marketing agency. It is one page of writing that turns a set of reactions into a set of decisions.
The one-page plan
Five questions. Answer them in your own words and you have something better than most small businesses operate with.
1. What is the job you want more of? Not "more work" — the specific job. "Full bathroom refits, not tap repairs." Everything else follows from this.
2. Who buys it, and what happens in their life just before they need you? The trigger event matters more than demographics. A house purchase, a failed inspection, a leak, a new baby, a season changing.
3. Where are those people already looking? Search, a neighbourhood group, a referral from an estate agent, a directory. Be honest rather than aspirational — ask your last ten customers how they found you.
4. What do you want them to do? One action. Call, book, or request a quote. Not all three.
5. How will you know it worked? One number, checked monthly. Usually enquiries for the job in question one.
Then pick three lanes and leave the rest
Marketing fails in small businesses mostly through spreading too thin. Six channels done badly return less than two done properly, and they cost more in attention, which is the resource you actually lack.
A common workable set for a local service business:
- Being findable when someone searches for the job. Your Google Business Profile, reviews, and a page per service. Slow to build, cheap to maintain, and it compounds.
- Making the most of the demand you already have. Faster response, better follow-up, asking past customers for repeat work. This is the cheapest growth available and it needs no advertising budget at all.
- One paid or outbound channel that suits your trade — local search ads, a partnership with a complementary business, a mailing to a specific neighbourhood after a job.
Everything else goes on a "not now" list. Writing it down as "not now" rather than leaving it undecided is what stops it creeping back in during a slow week.
Give it a rhythm you can actually keep
The plan that survives contact with a busy month is the modest one.
Weekly, fifteen minutes: ask for reviews from that week's completed jobs, and reply to any that came in.
Monthly, an hour: look at your one number. How many enquiries, from where, for which service. Compare to the previous month and to the same month last year, because seasonality will otherwise fool you completely.
Quarterly, half a day: re-answer the five questions. Has the job you want more of changed? Is a lane not returning anything? Kill it rather than defending it.
That is the whole system. Nine hours a year of thinking, and it will outperform reactive spending several times over.
Measure the thing, not the proxy
Almost every marketing report you will be shown measures a proxy: impressions, reach, clicks, followers, rankings. None of them pay wages.
Count enquiries and booked jobs, and where they came from. That requires two dull things: conversion tracking on the website, and asking every caller how they found you and writing the answer down. The second is low-tech, slightly annoying, and more accurate than most analytics for phone-driven businesses.
Be sceptical of anyone who reports on rankings without reporting on enquiries. Position three for a phrase nobody searches is not a result. It is a screenshot.
Questions people ask about this
What if I do not know which channel my customers use?
Ask the last ten. Not a survey — a question at the end of the job. "How did you come across us?" Ten answers will tell you more than any market research you could buy, and the pattern is usually obvious by number six.
How much should a small service business spend?
There is no honest universal percentage. Work it backwards: what is a job worth to you, how many extra jobs do you want next quarter, and what would you pay to get one? That gives you a ceiling per enquiry, which is a far more useful number than a share of revenue.
Should I do social media?
Only if the job you want more of is genuinely discovered there, and only if you will sustain it. For most trades, search and reviews return more per hour. For visual work — landscaping, renovations, med spa — before-and-after photos on social are a real channel worth committing to properly.
How long before any of this shows results?
Response and follow-up improvements show up in weeks. Search visibility takes months and compounds afterwards. Paid channels show up immediately and stop the moment you stop paying. Plan with all three timescales in mind so a slow month does not cause you to abandon the slow-burning work.
