Spreadsheet to Dashboard: A Singapore SME Case Study (9 Tabs, 3 Versions Named Final)
Yes, a Singapore SME can move off its master spreadsheet and onto a live dashboard in about three weeks, for a one-off cost in the low thousands and a running cost under S$50 a month. This case study covers a 24-person distributor in the west of Singapore that did exactly that in 2026: nine tabs, four people editing the same file, three copies named some version of final. The details below are anonymised with the client's permission, but the numbers are theirs.
What was the spreadsheet actually costing?
The company sold industrial consumables to about 180 trade accounts. One workbook held everything: open orders, stock on hand, outstanding invoices, a sales-by-customer tab, a pricing tab, and four more that had grown out of one-off requests from earlier years.
Nobody thought of it as a cost. It was just how the business ran. So we measured it for two weeks before touching anything:
- The operations manager spent roughly 40 minutes every Monday rebuilding the same five numbers for the owner.
- Month-end stock and sales reconciliation took a full day, because the stock tab and the accounting system disagreed and somebody had to decide which one was right.
- Three people had their own saved copy of the file, so quotes were being issued off prices that were between two weeks and four months out of date.
- Twice in the previous year, an order was shipped against stock that had already been promised to another customer. One of those cost an air freight charge of about S$900 to make good.
Added up, that is roughly 12 to 14 working days a year of pure re-typing and checking, plus the occasional four-figure mistake. That is the real number an owner should weigh, not the software licence.
Why not just fix the spreadsheet?
This was the first thing we looked at, and it is often the right answer. A spreadsheet is cheap, everybody already knows it, and no consultant is required to change a column heading.
It failed here for one reason: more than one person needed to be in it at the same time, and they needed to trust what they saw. A spreadsheet is a single-author document. Once four people are editing it and three of them keep a local copy, you no longer have one version of the truth, you have four opinions with the same filename. No amount of formula tidying fixes that.
Our rule of thumb for a Singapore SME: keep the spreadsheet if one person owns it and the rest only read it. Move off it when two or more people need to write to it, or when a number in it triggers money leaving the company.
What replaced it?
Deliberately, not much. We did not propose an ERP. We built the smallest thing that answered the five questions the owner asked every Monday.
- A MySQL database holding four tables: customers, products, orders, order lines. Nothing else in phase one.
- A simple internal web page for order entry, so the warehouse and the two sales staff write to the same place.
- One dashboard page showing the five numbers: orders today, open order value, stock below reorder level, invoices overdue past 30 days, and sales this month against last month.
- A nightly import from the accounting system for invoice payment status, so nobody re-types what the accounts package already knows.
- An export button that produces the old spreadsheet layout, because the external bookkeeper still wanted it. This mattered more than it sounds.
The nine tabs did not all survive. Four of them turned out to be reports nobody had opened in over a year. We archived them and nobody has asked.
How long did it take and what did it cost?
Three weeks end to end, broken down as one week of watching how the work actually happened, one week of build, and one week of running the spreadsheet and the dashboard side by side before switching off the old file.
The one-off cost was in the low thousands of Singapore dollars for a build of this size. The running cost is server hosting of under S$50 a month, with no per-user licence, which matters for a company that adds and loses warehouse staff through the year. For a Singapore SME considering this, a Productivity Solutions Grant claim may apply if you buy a pre-approved package instead of a custom build, which is a genuine trade-off: the grant reduces the cash outlay, a custom build fits the way you already work. Price both.
That parallel-running week is the part people cut to save money. Do not cut it. It is the only week where a mistake is free.
What changed in the first month?
The Monday morning rebuild went from about 40 minutes to zero, because the owner now opens a page instead of asking a person. That is roughly three hours a month back for the operations manager, which the company redirected to chasing overdue invoices.
Month-end reconciliation dropped from a day to about two hours. The stock and sales figures now come from the same order record, so there is nothing to reconcile, only to check.
The quieter win was pricing. One price list, one place, visible to everyone quoting. Two months in, the owner found a customer who had been quoted an old price for most of a year. That discovery alone was worth more than the build.
What went wrong?
Two things, and both are worth repeating because they are not unique to this client.
First, the historical data was dirtier than anyone expected. The same customer existed under four spellings, including one with a trailing space. We imported it as-is on day one and the dashboard showed four customers where there was one. Cleaning the master list took two extra days that were not in the plan.
Second, order entry was slower for the first fortnight. A web form with required fields is genuinely more friction than typing into a cell. One staff member went back to the spreadsheet quietly for a week. We only caught it because the order numbers did not tally. The fix was not a lecture, it was removing three fields from the form that nobody actually needed.
How does a Singapore SME start this?
Do not start with software. Start by writing down the five numbers you ask somebody for every week, and who they ask to get them. If those five numbers all come out of one spreadsheet that more than one person edits, you have the same problem this client had, and the same three-week fix is available to you.
Frequently asked questions
Do we need to replace our accounting system too?
No, and you should not try. The accounting system stayed exactly where it was. The dashboard reads payment status from it overnight and writes nothing back. Keeping the statutory records in the accounting package and the operational view in the dashboard is the cheaper and safer split.
Can we use Google Sheets or Power BI instead of a custom build?
Often yes, and it is the right starting point for a smaller company. A shared Google Sheet solves the multiple-editor problem for under 10 users and costs almost nothing. It stops being enough when you need enforced rules at the point of entry, such as refusing an order for stock that is already committed. That is a database job, not a spreadsheet job.
What happens if the person who built it leaves?
Ask for three things in writing before you start: the source code in a repository you own, the database credentials held by a director, and a one-page note on how to restore from backup. A build you cannot hand to the next supplier is a liability no matter how good the dashboard looks.
Ready to Transform Your Business?
Let Digital Perpetual help you automate, streamline, and grow.
Get Started with Digital Perpetual →