Integration & data 5 min read
Running several sites? How to compare them on the same numbers
When each site reports its own way, you cannot tell which one needs help. Here is how to get fair, like-for-like figures for every location.
By Digitice Admin
Every Monday, each site manager sends in their numbers. One uses the new till system. Another still uses the old one. A third sends a spreadsheet they built themselves. The leadership meeting then spends half an hour arguing about whose figures are right.
This has a real cost. Someone loses days each month copying numbers into one sheet. The figures arrive late. Worse, you may put money, stock or staff into the wrong site, because the comparison was never fair in the first place.
This article explains why multi-site numbers rarely line up, and how to fix it step by step. You do not need to be technical to follow it.
Why your sites do not compare fairly
The problem is rarely that one manager is wrong. It is usually that each site counts things a little differently. Common causes include:
- Different systems. Sites opened or bought at different times often run different tills, booking tools or stock systems. Each one reports in its own way.
- Different definitions. One site counts sales before refunds. Another counts them after. Both call it "sales".
- Different time periods. One site's week ends on Saturday. Another's ends on Sunday.
- Manual changes. Figures are typed, corrected and adjusted by hand before they are sent in. Nobody can see what changed.
- Different sizes and hours. A large site open seven days will always beat a small site open five. That does not mean it is better run.
Each difference is small. Together, they make the comparison unreliable.
Start with the decisions, not the data
It is tempting to start by collecting every number you can. This usually ends in a large report that nobody reads.
A better start is to list the decisions you make about your sites. For example:
- Which site needs help this month?
- Where should we move staff or stock?
- Which manager is doing something the others should copy?
- Is the newest site on track?
Then ask which figures you need to answer each question. Most businesses find that five to eight measures cover almost everything. Measures you never act on can wait.
Agree one definition for each measure
A metric definition is a short, written rule that says exactly how a number is worked out. It sounds dull. It is the most useful step in the whole process.
Here are some typical examples:
| Measure | Where sites often differ | Example of an agreed rule |
|---|---|---|
| Weekly sales | Before or after refunds and VAT | After refunds, excluding VAT, Monday to Sunday |
| Staff cost | Agency staff included or not | All paid hours, including agency, at actual cost |
| Customer numbers | Transactions or people | Number of completed transactions |
| Stock loss | Write-offs counted or not | Recorded write-offs plus count differences |
Write each rule down and agree it with the site managers. If they help set the rules, they are far more likely to trust the results.
Options for bringing the data together
Once the rules are clear, you need a way to collect the numbers. There are three common options.
| Option | Good for | Main drawback |
|---|---|---|
| Shared spreadsheet, filled in by hand | Two or three sites, few measures | Slow, easy to change by mistake, always late |
| Connecting tools with simple automation | Moving data between a few tools | Hard to manage as sites and systems grow |
| Pipelines into a single reporting database | Many sites, many systems, regular reporting | More set-up work at the start |
A data pipeline is an automatic feed. It copies data from a system, such as a till or booking tool, on a regular schedule. A reporting database is one central place where all those feeds land and where your agreed rules are applied.
The spreadsheet is fine while you are small. Once you have several sites on different systems, it tends to break. If your main need is to move data between tools, our page on integration and automation explains the options. If you need reliable reporting across many sites, a reporting database is usually the safer choice.
Make comparisons fair, not just equal
The same number for every site is not the same as a fair comparison. A busy city site and a quiet town site will never look alike on raw totals.
To compare fairly, look at figures relative to size or effort, such as:
- sales per opening hour
- sales per staff hour
- sales per square metre
- this year against the same period last year, for each site
New sites need special care. They often take months to settle. Compare them against how your other sites did in their early months, not against mature sites.
Keep the main view simple. A dashboard is a screen that shows your key figures and updates itself. A good one shows each site side by side, with a short note on what "good" looks like.
A practical way to start
You do not need to fix everything at once. A steady approach looks like this:
- Pick three decisions you make most often about your sites.
- List the figures behind each decision and where they come from today.
- Agree the definitions with site managers and write them down.
- Build the feeds for the most important measures first.
- Add checks, such as warnings for missing days or duplicate records, so problems show up early.
- Run the new report alongside the old spreadsheet for a month. Look into any gaps.
- Retire the spreadsheet once people trust the new figures.
This order matters. If people do not trust the numbers, they will go back to their own spreadsheets. Running both side by side for a short time builds that trust.
There is a further benefit. Clean, well-defined data is also what AI tools need to give sensible answers. If you plan to use AI later, this work prepares the ground. You can read more about how we approach this on our data and analytics page.
Your next step
Before your next leadership meeting, write down the three questions about your sites you most need answered each week. Then note where each answer comes from today, and how long it takes to get. That short list will show you where the biggest gaps are.
If you would like a second view on it, you can get in touch with us and we will talk it through with you.
Questions
Frequently asked questions
Do all our sites need to use the same system before we can compare them?
No, you can compare sites on different systems as long as their data is brought together and the same rules are applied to it. Data pipelines pull figures from each system, and the reporting database turns them into one consistent set of numbers. Moving every site onto one system can help later, but it is not needed first.
How many measures should a multi-site dashboard show?
Most businesses need only five to eight core measures. Start from the decisions you make each week and include only the figures that help you make them. Extra measures can be added later if people actually use them.
Can we keep using our spreadsheet while the new reporting is built?
Yes, and we recommend it. Running the old spreadsheet alongside the new reports for about a month lets you check the figures and explain any gaps. Once people trust the new numbers, the spreadsheet can be retired.
Will this work help if we want to use AI later?
Yes, because AI tools depend on clean, clearly defined and well-organised data. Agreed definitions, checks for errors and controlled access make it much easier to connect an AI assistant to your business data safely.