Picking the trade and the town

This is the highest-leverage hour in the whole course and it costs nothing. A perfectly built site aimed at the wrong trade in the wrong town cannot be rescued by effort later. A mediocre site aimed well will still produce calls.

The trade: what one job is worth to the buyer

A call is worth a fraction of a job, so the ceiling on what anyone can pay you is set by their ticket. A trade where a booked job is two hours of labour will never support much. A trade where one job is a multi-day install can. You do not have to guess at the number — later you will ask the buyer directly what a closed job is worth to them. For now, sort your candidates by instinct against five filters.

• Urgency. Is this searched at the moment of need, or researched over three weeks? Urgent searches turn into phone calls. Considered ones turn into browsing.
• Ticket. Is one completed job worth real money to the business that does it? If not, there is nothing for you at the end of the chain.
• Physical. Does someone have to show up in person? If the work can be done from anywhere, you are competing with the entire internet instead of one town.
• Repeatable. Does this get searched more than a handful of times a week in a town this size? A trade people need once a decade will not fill a month.
• A buyer exists. Are there businesses in that town doing this work, and are they small enough to want the phone to ring more?

That last filter is where beginners flip the logic. It feels clever to pick a service nobody in the town appears to offer. But you are not the one doing the work. If there is no established business in the trade, there is nobody to rent the calls to, and you have built a very good sign pointing at an empty lot.

The town: big enough to search, small enough to enter

In a large metro, the top of local results is held by businesses with a decade of reviews, an established profile and somebody on staff whose entire job is this. You will not displace them with a new domain and a weekend. In a village of eight hundred people, the searches you need are not happening at all. You are looking for the band in between, and you find it by looking rather than by rule.

• Does the town have a name people actually use when they search? Some suburbs do not — everyone types the bigger city's name instead, and you are back in the metro fight.
• In the map results for your phrase, are the profiles carrying three reviews or four hundred? Write the number down. That number is your difficulty score and it is more informative than any tool.
• Do several of the businesses in that trade have no website at all, or one that has plainly not been touched in years? That tells you what the ranking bar looks like and it tells you who your buyer might be.
• Can you get there? Not essential, but the location pages you write about a town you have driven through are noticeably better than the ones you write about a town you have only seen on a map.

Where the county data comes in

Scoutway compares how many businesses in an industry exist in your county against the national rate per resident, using two US Census datasets together — County Business Patterns for firms with paid employees and Nonemployer Statistics for sole proprietors — and returns one of three verdicts for that county and industry: underserved, average or saturated. For this business it is a useful input and it is not an answer. Knowing precisely what it measures is what makes it useful rather than misleading.

• It measures supply. It counts how many businesses of a type exist per resident. It has never observed a single person wanting the service, and neither has anything else in this product.
• It counts sole proprietors as well as firms with staff, and in most of these trades the sole proprietors are the larger half by a wide margin. That matters here specifically, because a one-truck operator is precisely the kind of business that buys leads. Where the sole-proprietor half cannot be measured for a category, the product shows no verdict at all rather than an employer-only one.
• The industry categories are coarser than your trade. A single Census code can cover several jobs that have nothing to do with each other, so the count is of a category your trade sits inside, not of your trade.
• It is a county. The town you are targeting is a fraction of it, and the count cannot be broken down below the county line.
• Very small counties are deliberately excluded, because a ratio built on two businesses is noise rather than signal.

So read it sideways rather than straight. A saturated verdict means a lot of that category per resident: a longer list of possible buyers, and probably more established sites to outrank. An underserved verdict means fewer of them on the books, which is a thinner buyer list — and it is not a promise that the search results will be empty, because a county count says nothing about who has bothered to build a website in the one town you picked. Neither verdict is a green light or a red one. Both of them shrink the guessing, and shrinking the guessing is all any data can honestly do for you here. The search results you looked at in the last section are the closer evidence; the county count only tells you whether it is worth looking.

Do this for three candidate pairs, not one. Three is enough for the differences to become obvious and few enough to finish today. You are looking for the pair where the trade passes the five filters and the top three map results are carried by profiles you could realistically catch — and where, when you run the search, you can already picture the business you would call.