From three stores up, the problems change: the same item at different prices, stock sitting in the wrong branch, and a daily report nobody acts on. What to fix first.
Key takeaways
- Price needs a single source; a store updating its own is a source of wrong prices at the till.
- Stock in the wrong branch is a lost sale, not merely surplus.
- A transfer between stores with no document is the common cause of count gaps.
- A daily report is worth something only if it contains exceptions rather than totals.
- One measure per store beats ten that nobody reads.
One store runs on memory; three do not. From that point three new problems appear that did not exist before: the same item at different prices, stock missing in one branch and sitting in another, and a daily report that goes out and is acted on by nobody.
Pricing: one source, not three
At a small chain, a price change is usually made at the store's till. That is convenient, and within two months it produces three states: a different price between stores, a till price that is not what the shelf says, and a promotion that ended in one store and continues in another.
The third is the expensive one, because it persists. The first produces an immediate complaint; a promotion nobody took down keeps eroding margin quietly until somebody notices.
What is needed is that price be set in one place and pushed to the stores, and that a local override be an exception that is recorded. No expensive system is required - what is required is a decision about who updates, and a weekly check that till prices match.
Stock between stores
| The situation | What actually happens | What is better |
|---|---|---|
| Item missing at store A, present at B | A lost sale | A transfer, with a document |
| A customer asks and it is out | They are sent to look themselves | An order to the store with a date |
| Surplus in one store | Sits until end of season | A proactive transfer based on sales |
| A transfer that happened | Was never recorded | Two counts that do not reconcile |
| Counting | Once a year | Cyclical, by category |
The fourth row is the root of most of the table. Once goods move between stores without being recorded, both stores report wrong stock - one over and one under - and every decision resting on those numbers is wrong too.
The daily report: why nobody reads it
The classic report sends end-of-day sales by store, transaction count and average basket. That is correct information and useless, because it does not say what to do tomorrow.
A report that gets read is an exception report: what deviated from expectation, not what happened. Five rows suffice - a store with an unusual drop, an item that ran out, a promotion that is not moving, a till variance above a threshold, and a count that was not done. If there are no exceptions the report is short, and that is a good sign rather than a problem.
The practical difference is that an exception report is read in thirty seconds and produces an action, while a totals report is read in thirty seconds and produces an impression. The broader side of till controls is covered in cash handling controls at the till.
What to measure per store
Choosing measures at a small chain tends to sprawl, and that is what kills their use. Four are enough:
- Sales against the same period last year.
- The share of top items that went out of stock.
- Count variance as a percentage.
- Labour cost as a percentage of sales.
The second is the one people tend not to measure and the most meaningful. A top item out of stock is a sale that was guaranteed, and unlike the other measures it is immediately fixable - usually by a transfer from another store.
Why is the third store the turning point?
With two stores everything still fits in one head: one manager knows both, and every exception reaches them directly. At three that breaks, because nobody is present at all three, and communication moves into phone calls.
What changes is not the volume but the need for definition: who sets price, who approves a transfer, and who owns counting. At two stores that definition is unnecessary; at three it is the difference between managing and firefighting.
It is also the point at which it is worth stopping managing in WhatsApp. Not because it does not work, but because decisions get swallowed in it - and when a store manager asks what was agreed on a price, the answer has to live in one place.
Sources
Frequently asked questions
Should each store count for itself?
Yes, but on a central schedule rather than at the store's discretion. A cyclical count by category - one category a week - spreads the work and produces a continuous figure instead of an annual event.
How do you decide which store to transfer from?
By sales rather than by quantity. A store with ten units selling two a week is the right source; a store with eight selling five is not, even though it holds less.
What do you do when a store manager changes a price without approval?
First find out why - in most cases it is a response to a competitor or a shelf error rather than a bypass. What has to change is that there be a fast route to request a change, because without one the bypassing will continue.
Do all stores need the same till system?
It helps a great deal and is not a condition. What is a condition is that the data reach one place in the same format - without that, any comparison between stores is manual work somebody will eventually stop doing.
Keep reading
Related service
Inventory & Purchasing
SKU-level stock, reorder rules and an approval flow that leaves a record.
About the author
Yehonatan Saadia
Freelance automation, web & MVP developer
I'm Yehonatan Saadia, a senior developer who builds business automation, custom websites, and MVPs for small and mid-sized companies across the US, Europe, and Israel. These guides come from real client work, not theory.
Work with meHave a project like this?
Tell me what you're trying to automate or build and I'll tell you the fastest reliable way to ship it.