Business

Choosing a Shop Location: Counting Instead of Guessing

K By Kev 20 November 2026 9 min read
Share
Business guide

Location is the decision you cannot fix later. Pricing, range, hours and service can all be adjusted after opening; the street cannot. And yet it is routinely chosen on a feeling about a place, often formed on a single visit at one time of day, which happened to be the time of day the street was busy. The alternative is not expensive research. It is a notebook, four visits and some arithmetic.

Quick answer

Count, do not estimate. Stand at the site at four different times across a week with a tally, count people passing and how many go into neighbouring shops, and work backwards to the sales the location could support. An hour of counting beats any amount of opinion about a street.

Key takeaways
  • Count, do not estimate. Four fifteen-minute counts at different times beat any amount of opinion about a street.
  • Traffic is necessary and not sufficient: what the traffic is doing decides whether it is traffic for your shop.
  • Work backwards from the rent to the customers per day the site must deliver, then compare that with what you counted.
  • A cheaper site with a third of the traffic can cost more per customer than a dearer one.
On this page▾
  1. What makes a location good for your shop specifically
  2. Testing a site properly
  3. How location decisions go wrong
  4. A worked example: the cheaper site that was not
  5. Checking the forecast against what happens
  6. Frequently asked questions

What makes a location good for your shop specifically

Foot traffic is necessary and not sufficient. A busy pavement outside a matatu stage carries people moving with purpose towards somewhere else, and a shop selling things people buy on impulse does well there while one selling things people plan for does not. The same pavement is a good location for one business and a poor one for another.

The second factor is what the traffic is doing. People walking home in the evening buy differently from people walking to work in the morning: the evening flow buys dinner, the morning flow buys airtime and a newspaper. A site whose peak is the wrong direction for your range is a weak site at any volume.

The third is competition, and it cuts both ways. Three shops selling the same thing on one street can mean a saturated market or an established destination that customers come to specifically. Which of the two it is can usually be determined by watching whether people arrive, compare and choose, or simply enter the first one.

Want to see it running in your own business first?

Sign Up

Testing a site properly

All of this is free and takes about four hours spread over a week. It is the highest-return work available before signing a lease.

  1. 1

    Step 1: Count foot traffic at four different times

    Morning rush, midday, evening rush, and a weekend afternoon. Fifteen minutes each with a tally, counting people passing in each direction. Multiply out carefully and sceptically. A fifteen-minute count is a sample, not a measurement, and four samples are much better than one.

  2. 2

    Step 2: Count conversions, not just passers-by

    Pick the neighbouring shop most like the one you intend to open and count how many of the people passing actually go in. That ratio, applied to your traffic count, is the closest thing to a sales forecast you can get for free.

  3. 3

    Step 3: Estimate basket size from what you can see

    Watch what people carry out. A shop whose customers leave with one small item has a very different average basket from one whose customers leave with a bag, and your break-even depends on the basket as much as the count.

  4. 4

    Step 4: Work backwards from the rent

    Take the monthly fixed cost, divide by your expected gross margin, and you have the sales the site must produce. Divide that by 26 trading days and by your estimated basket, and you have the customers per day required. Compare with step two. If the required number is above what you counted, the site cannot work at that rent regardless of how much you like it.

  5. 5

    Step 5: Ask the neighbours what changed

    Shopkeepers on a street will tell you what happened last year: a stage that moved, a road that closed, a market that relocated, a competitor who left and why. This is information that exists nowhere else and costs a conversation.

  6. 6

    Step 6: Find out what is coming

    County offices, the landlord’s other tenants and the neighbours all know about planned roadworks, a new development or an incoming competitor. You are buying the next three years of the street, and the last three are only a proxy.

How location decisions go wrong

Visiting once, at the busy hour

Every street has a peak and most have a long flat afternoon. A single visit at the peak produces a confident and wrong picture, and it is the most common error in this list.

Mistaking traffic for the right traffic

A thousand people hurrying to a matatu are not a thousand potential customers for a shop selling planned purchases. Count what the traffic is doing, not only how much of it there is.

Choosing on rent alone

A cheap site with a third of the traffic is more expensive per customer than a dear one. The comparison that matters is rent against expected sales, not rent against rent.

Assuming you will create your own traffic

A destination shop on a quiet street is possible and it is a marketing business as much as a retail one. If that is the plan, it should be a deliberate plan with a budget, not a hope attached to a cheap lease.

Ignoring the side of the street

On roads people cross reluctantly, one side can carry most of the trade. Count both sides separately; the difference is often larger than anyone expects.

A worked example: the cheaper site that was not

Worked example

Two units are available for a general shop. Site A is KES 55,000 a month on a main road. Site B is KES 28,000 a month on a side street 150 metres away.

Counting at four times, site A averages around 420 people an hour past the door across trading hours; site B averages about 95. The neighbouring comparable shop at A converts roughly 4% of passers-by; at B, a similar shop converts about 9%, because the people on a side street are more likely to be there deliberately.

That gives A roughly 17 customers an hour and B roughly 8.5. Over a 12-hour day, about 200 against 100. At an observed average basket of KES 140 and 26 trading days, A implies around KES 728,000 of monthly sales and B around KES 364,000.

Now the rent test. At a 19% gross margin, A needs sales of about KES 289,000 a month to cover a KES 55,000 fixed cost; B needs about KES 147,000 to cover KES 28,000. Both clear their hurdle on these estimates, but not by comparable amounts. A produces roughly KES 138,000 of gross profit a month, leaving about KES 83,000 after the rent. B produces about KES 69,000, leaving roughly KES 41,000.

The cheaper rent is the worse deal here, and nothing about standing on either street would have told the owner that. The counting did. It is also worth saying plainly that these are estimates built on fifteen-minute samples, and the right response to a close result is more counting rather than more confidence.

Business impact

Running a shop on memory and paper leaves money on the table: missed sales, stock you cannot account for, and receipts KRA will not accept.

Veira records every sale, tracks stock and keeps you eTIMS-compliant automatically, so the numbers look after themselves.

Checking the forecast against what happens

The arithmetic in step four needs a gross margin and an average basket. For a first shop those are estimates; for a second, they are figures you already have, which makes site selection dramatically more reliable the second time.

Veira records average transaction value and margin as the shop trades, so the forecast you built before signing can be compared with what the location actually produced. That comparison is what makes the next location decision better than this one.

Reading a street, and knowing which side people walk on, is local knowledge that no system holds. The counting is yours to do and it is the part that matters most.

Frequently asked questions

How do I measure foot traffic for a shop location?
Stand at the site with a tally at four different times: morning rush, midday, evening rush and a weekend afternoon. Fifteen minutes each, counting both directions and both sides of the street separately. Treat the result as a sample rather than a measurement.
How much foot traffic does a shop need?
There is no universal number, because what matters is traffic multiplied by the rate at which it converts and by basket size, measured against your fixed costs. Work backwards: fixed cost divided by margin gives required sales, divided by trading days and basket gives required customers per day.
Is a cheaper location usually better?
Often not. A site with a third of the traffic can cost more per customer than one at twice the rent. The comparison that matters is rent against expected sales, not rent against rent.
Is it bad to open near competitors?
Not necessarily. Several similar shops on one street can mean a saturated market or an established destination customers travel to. Watching whether people compare before choosing, or simply enter the first shop, usually tells you which.
Does the side of the street matter?
More than most people expect, especially on roads pedestrians cross reluctantly. Count each side separately rather than counting the street as a whole.
What should I ask the neighbouring shopkeepers?
What changed in the last year or two: a matatu stage that moved, a road closure, a market relocation, a shop that closed and why. This information exists nowhere else and costs only a conversation.
Can a shop create its own foot traffic?
Yes, and it becomes a marketing business as much as a retail one. If that is the plan it needs a budget and a deliberate strategy, not just the saving on a cheaper lease.

Location is the one decision you cannot adjust after opening, and the cheapest way to get it right is a notebook and four visits. Count traffic, count conversions at a comparable shop, and work backwards from the rent to the customers the site must deliver. See how Veira tracks basket size and margin so your next site decision rests on your own figures.

Need help, or want to talk it through first? Chat with us on WhatsApp
Terms explained

Keep reading

See all Business guides

Veira for your business

Browse Veira by business type · Who we serve