You need a cleaner for the office, two electricians who can quote on a small fit-out, and someone to fix the sign above the door. The instinct is to open a dozen tabs, collect forty names, and then stall. The list gets longer, the decision gets harder, and a week later you are still searching.

A shortlist does not need to be exhaustive. It needs to be defensible: five names you can explain to a colleague, with contact details that work. That is an afternoon's job, provided you time-box it and resist the urge to keep looking. Here is a workflow that fits between lunch and the end of the day, with room for coffee.

Write the brief before you open a browser

Twenty minutes here saves an hour later. Before you search anything, write down what you are actually buying. Not "a plumber" — something closer to "a plumber who can attend a ground-floor retail unit within 10 miles, handle a leak behind a fitted cabinet, and invoice a limited company."

Four lines are enough:

  • The job, in plain words. One sentence, no jargon.
  • The service area. A postcode, a district, or a radius in miles. Be strict here; it eliminates more candidates than any other filter.
  • Non-negotiables. Insurance, licensing, out-of-hours availability, minimum job size they will take.
  • The timeline. Do you need someone this week, or are you planning for next quarter? Your answer changes who is worth calling.

Keep the brief visible while you search. It is the difference between a list of vendors and a list of vendors who could plausibly work for you.

Sweep the listings in one direction

Set a timer for 45 minutes and work through sources in a fixed order, so you are not circling back. Start broad, then narrow by sector. General directories and map results give you volume and opening hours; sector associations and trade bodies give you fewer names but a higher hit rate.

Sources worth a pass, depending on the trade:

  • Map and search listings for the category plus your district — check the results on the map as well as the ranked list, since some businesses sit just outside the boldest results.
  • Business directories, including regional ones that cover your area rather than national aggregators with thin records.
  • Trade associations and licensing registers where the sector has them.
  • Chamber of commerce and local business networks, which often list members by category.
  • Neighbourhood groups and local forums for names people mention unprompted — treat these as leads, not references.
  • Your own inbox and invoices. Someone in the building has hired a locksmith before.

Capture each candidate as you go: business name, phone, website, address, and the source you found them in. Do not judge yet. Judging and searching at the same time is how an afternoon disappears.

Keep a working table, not a pile of tabs

Fifteen minutes to build a spreadsheet with one row per vendor and these columns: name, category, area, phone, email, website, source, date checked, and a free-text notes field. Add a status column — new, maybe, no, call — and a one-to-five score if you like ranking things.

Two practical notes. First, record the date you checked each listing; directory data goes stale, and a phone number verified in March is a different asset in September. Second, copy the email address into the sheet even if you plan to call. Outreach always ends up needing it.

Cut the list in three passes

Set aside an hour. The goal is to get from twenty-five names to ten, and the work is mostly deletion.

Pass one: geography and capability

Can they physically do the job at your location? Delete anything outside your service area, anything whose listed specialism is clearly different from yours, and anything that only serves a customer type you are not.

Pass two: signs of an operating business

Look for evidence that someone is answering the phone and turning up to jobs: recent reviews, a website with a real address, photographs of completed work, responses to critical reviews that sound like a person rather than a template, opening hours that match the trade.

Pass three: the ones worth a call

Stop at eight to ten names. If you are left with fifteen, cut the five weakest on evidence, not on gut feeling. Keep the gut notes in the spreadsheet — they are useful later — but let the visible signals make the decision.

A few things justify a fast no:

  • A listing with no address and no phone number, only a contact form.
  • The same business appearing under three names with different numbers.
  • A cluster of glowing reviews posted in the same week, with nothing before or after.
  • No indication of coverage area, so you cannot tell whether they will come to you.
  • A website that has not been touched in years while the listing claims active service.

Spend the phone time

Budget forty minutes for six calls. Phone beats email at this stage because you learn things a form cannot tell you: whether someone picks up, how they handle a basic question, whether they ask you anything back.

Keep a short script so the calls stay comparable:

I'm looking for [service] at [location] for [rough scope]. Four quick questions: do you cover that area, what is your earliest availability, who would actually do the work, and can you send a written quote by email?

Write down the answers immediately, plus the time of the call and the name of the person you spoke to. A vendor who answers on the second ring and asks two sensible questions about access and timing is telling you something useful. So is one who cannot name a start date. If a call goes to voicemail, leave a message and mark it for a single follow-up the next morning — no more than one.

Write the shortlist so outreach is easy

Twenty minutes to turn notes into a document someone else could pick up and use. Aim for three to five entries. Each one needs:

  1. The business name and the name of your contact.
  2. Phone and email, verified today.
  3. One line on why they made the cut — "covers the postcode, insured, can start in two weeks."
  4. One line on what you still need to confirm, such as pricing structure or references.
  5. The next action and a date: "request written quote by Friday."

Keep it to a page. A shortlist that runs to three pages is a research file, and nobody acts on a research file.

Finish at the finish line

The afternoon works because it has an end. Search block, filter block, call block, write-up. When the timer goes, stop looking, even if you suspect there is a better plumber somewhere in the next district. There usually is — and you can find them next quarter if the first five do not work out.

Two habits make the next round faster. Note the date on everything you verified, and keep the discarded names in a separate tab with the reason they were cut. Half of them will be worth a second look in six months, and you will have the reasoning already written down.

Photo: StartupStockPhotos / Pixabay