Still running your business on spreadsheets? A 5-question test for custom software

You should replace Excel with custom software when the spreadsheet keeps causing the same costly problems: people overwriting each other, data typed into several places, rules nobody can enforce, one person holding it together, and reports that take days. If fewer than four of those are true, fix the spreadsheet first. A focused custom tool starts at ₹2,50,000 at Gigmoz.

That is the honest short version. Spreadsheets run many Indian businesses well, from distributor order books to clinic appointment lists. They are cheap, flexible and everyone knows how to use them. The trouble starts when the business grows around the sheet faster than the sheet can grow with it.

This guide gives you a five-question test, the cheaper fixes worth trying first, and what a first custom release looks like and costs if you do build. It is written for owners and operations heads, not developers. Many readers will finish it and decide not to build yet, and that is a good outcome.

The five-question test

Answer each question for the one spreadsheet that causes the most trouble, not for the business as a whole. Be strict: a yes means it happens most weeks, not once a year.

1. Do several people edit the same sheet, and lose each other’s work?

What a yes looks like: the order sheet lives on a shared drive or in a group email, and someone regularly asks “which version is the latest?”. Two people sort the same tab at once and rows end up against the wrong customer. A salesperson keeps a private copy because the shared one is “always being changed”.

Why it matters: a spreadsheet is designed around a person working on a grid. Shared editing in Google Sheets or Excel online helps, but it does not stop someone deleting a column that ten other formulas depend on. When the sheet is the single record of your orders, stock or leads, one bad edit is a business problem, not a typing mistake.

2. Is the same information typed into more than one place?

What a yes looks like: an order is taken on WhatsApp, typed into the order sheet, typed again into Tally as an invoice, and typed a third time into a dispatch register. Customer details exist in four slightly different spellings across those places.

Why it matters: every re-typing is time, and every re-typing is a chance for a wrong quantity, price or GST number. Re-entry is also where people quietly stop updating one of the copies, so the copies drift apart. This question is often behind a real case for software, because removing re-entry saves time every day.

3. Does the work need rules the sheet cannot enforce?

What a yes looks like: discounts above a limit need a manager’s approval, but anyone can type any discount. Branch staff should only see their own customers, but they can see everyone’s. An order should not move to “dispatched” until payment is confirmed, but nothing stops it.

Why it matters: spreadsheets can lock cells and restrict values, but they cannot easily say “this person may do this step, in this order, after that person has done theirs”. When your controls depend on everyone remembering the rules, they fail on the busy days, which are the days that matter.

4. Does one person hold the sheet together?

What a yes looks like: there is a person (often the owner, an accountant or a long-serving coordinator) who built the formulas, knows which tabs not to touch, and fixes it when it breaks. When they are on leave, the monthly report waits.

Why it matters: this is key-person risk. The sheet works, but only while that person is available. It also limits growth, because adding a branch, a product line or ten more staff means more work for the one person who understands it.

5. Are the reports slow, or argued over?

What a yes looks like: the weekly sales or stock report takes someone half a day of copying, filtering and pivoting. Two managers bring different numbers to the same meeting, each from their own copy. Nobody trusts the “pending payments” figure without checking it by hand.

Why it matters: the point of keeping records is to make decisions with them. If producing a number is slow, you look at it less often. If two copies disagree, meetings turn into reconciliations instead of decisions.

How to read your score

The score is a guide, not a formula. A single yes to question 2 can justify an integration on its own if the re-typing is into your accounts and happens hundreds of times a month.

Try these cheaper steps first

Before you pay anyone to build software, try the cheapest fix that matches your biggest problem. Each of these is enough for some businesses, and each one teaches you what you actually need, which makes any later build cheaper and more accurate.

A better spreadsheet structure

Most troubled spreadsheets are troubled because of how they are laid out, not because spreadsheets are the wrong tool. The fixes are free:

• One row per record and one tab per kind of thing: customers on one tab, orders on another, products on a third. No merged cells and no totals in the middle of data.
• An ID for every record, such as a customer code, so you look things up by code instead of by name.
• Inputs, calculations and reports kept apart, so nobody types over a formula.
• One shared file, not copies sent around by email.

If your problem was mainly slow reports or broken formulas, this may be all you need.

Forms and data validation

Both Excel and Google Sheets let you restrict what people can type into a cell. Microsoft describes Excel’s data validation as a way to “restrict the type of data or the values that users enter into a cell”, with drop-down lists and error messages when someone enters something invalid (Microsoft Support). Google Sheets offers in-cell drop-down lists that can reject entries not on the list or simply show a warning (Google Docs Editors Help).

Add a form in front of the sheet and staff never touch the grid at all. Google Forms can save every response straight into a Google Sheets spreadsheet (Google Docs Editors Help). For field staff logging visits or orders from a phone, that alone removes a lot of bad data.

No-code tools

No-code and low-code platforms such as Airtable, Google AppSheet and Zoho Creator let a non-developer build a simple app on top of a table: a form, a list, a few statuses and a dashboard. They are a good fit for a small team with simple rules. They get harder to manage as permissions, integrations and the number of users grow, and their per-user subscriptions can add up. Check each vendor’s current plans and limits before you commit.

An off-the-shelf product

If the spreadsheet is doing a job that thousands of businesses do the same way, buy the product built for it. Accounts belong in accounting software such as Tally. A standard sales pipeline often fits a subscription CRM. A single shop with ordinary billing usually fits a standard billing or POS product. If you are weighing a CRM specifically, our guide to custom CRM cost in India does the three-year sum against a subscription.

If you answered yes to fewer than four questions, or you have not tried any of the steps above, you should not build custom software yet. A month with a tidier sheet or a trial of an off-the-shelf tool will either solve the problem or show you exactly what to build.

When custom software is the right answer

Custom software earns its cost when the cheaper steps have been tried and the same problems keep coming back. In practice that usually means one or more of these:

• Your process is your advantage, and every off-the-shelf tool asks you to change it.
• The real work is moving data between systems, such as orders from WhatsApp into Tally, or a website enquiry into a sales pipeline and then into an invoice.
• Access and approval rules matter: branches, dealers, field staff and managers each see and do different things.
• Per-user subscription fees grow with headcount, so people stay outside the tool and keep their own sheets.
• The workarounds already cost money every month: exports, re-typing, extra tools stitched together by hand.

Custom does not mean everything at once. The goal of a first release is to replace one spreadsheet and the workflow around it, used every day by the people who used to edit the sheet. Our custom software development work starts from that one workflow, and connecting it to your other tools is API integration work scoped alongside it.

What a first custom release looks like

A first release replaces the one spreadsheet that scored highest on the test, and nothing else. For example, a distributor replacing an order tracker might get:

• One kind of record, here orders, with the fields the team actually fills in, and linked customer and product lists.
• Statuses that match the work, such as received, confirmed, packed, dispatched and paid, with a person responsible at each step.
• Two or three roles, for example sales, dispatch and owner, each seeing and changing only what they should.
• A few rules the sheet could not enforce, such as no dispatch before approval, or discounts above a limit needing a manager.
• A one-time import from the existing spreadsheet, after cleaning.
• Three or four reports that replace the ones someone used to build by hand.
• An export to Excel, because people will still want to analyse a copy.

What it usually leaves out is just as important: integrations with other software, a native mobile app, and every other spreadsheet in the business. Those come in later releases, once the first one is in daily use and you have learnt what is worth adding.

What it costs

The numbers below come from the software bands on our pricing page. They are starting estimates, confirmed only after a scoping call, and fixed wherever the scope allows.

Three things sit outside the build price:

• Hosting and third-party services such as WhatsApp messaging, SMS or email, billed by their providers and listed separately.
• Support and change. Our software support retainers typically run ₹25,000 to ₹75,000 a month, depending on how much you expect to change.
• Your team’s time for discovery, testing and cleaning the data before import. It is the cost most often forgotten.

You own the code. Full IP transfer on final payment is part of our standard engagement. Payments are split across milestones, and anything outside the agreed scope is quoted before it is built.

Why first releases go wrong

Spreadsheet replacements tend to fail for the same few reasons, and all of them are avoidable:

1. Trying to replace every sheet at once. The project grows, the launch slips and staff keep the old sheets “just in case”. Start with one.
2. Copying the spreadsheet screen for screen. The sheet’s layout reflects its limits. Build around the work, not around the old columns.
3. Skipping the people who do the work. The owner knows what the report should say; the coordinator knows why column K exists. You need both.
4. Importing dirty data. Duplicates and inconsistent names come straight across unless someone cleans them first.
5. Running both systems for months. Agree a cut-over date and a rule that the new system is the record from that day.

How to prepare for a scoping call

The fastest route to an accurate quote, or to an honest “do not build yet”, is to bring evidence of how the work happens today. Bring three things.

The spreadsheet itself. A copy with real rows if you can share them, or a sample with names masked. It shows the true data model, the workarounds and the mess, which no feature list can.

The people who touch it. List everyone who adds, edits, checks or reads the sheet, and what each of them does with it. If you can, bring the person who maintains the formulas to the call.

The reports. The three to five numbers or reports that management reads from the sheet, and how long each takes to produce today.

Then check these before you book:

• I have scored the five questions for one specific spreadsheet.
• I have tried, or ruled out with a reason, a tidier sheet, forms and validation, a no-code tool and an off-the-shelf product.
• I know which other systems the data must reach, such as Tally, WhatsApp or the website.
• I know who will test the first release and who decides when it is ready.
• I have a rough cut-over date in mind.

The scoping call is free and takes 30 minutes. We usually send a written proposal within three working days, and if a spreadsheet fix or an off-the-shelf tool is the better answer, we will say so on the call.