A small business should move from spreadsheets to a business web app when 5 signs appear: several people entering data in one file, the same data duplicated across sheets, broken formulas nobody dares to fix, no record of who changed what, and reports assembled by hand from several places. The sensible path is not a "rebuild from scratch" but 4 stages: standardize the data, set up a shared database, build entry screens for the real workflows, then automate and report. Each stage takes 3 to 6 weeks, and the business keeps working normally throughout.

Spreadsheets did their job very well

Excel and Google Sheets are the best startup tools ever made: free or cheap, usable by everyone, editable on the spot. Running inventory, orders, receivables and timesheets on spreadsheets for years is normal and correct.

Trouble starts when the business outgrows the spreadsheet. Not because there is too much data, but because many people and many processes touch the same file.

5 signs it is time to move

  1. Several people entering data in one file. Overwriting each other, "who deleted my row", locking the file so people take turns.
  2. The same thing in several sheets. A customer exists in the sales file, the receivables file and the care file; fix the phone number in one and the other two are wrong.
  3. Formulas nobody dares to touch. The person who built the file has left, the formula spans 6 sheets, and a #REF! stops the whole department.
  4. No change history. Stock was 120 yesterday, 80 today, nobody knows who changed it, when or why.
  5. Reports assembled by hand. Every Friday someone spends half a day copying numbers from 4 files into a summary.

Two signs means start planning. Four or more means you are paying a hidden cost every day; see the hidden costs of manual operations.

The common mistake: buy packaged software or rebuild everything

Two typical reactions, and both tend to fail:

Buying packaged software (ERP, off-the-shelf management tools): fits if your processes match the industry standard. But the reason small businesses use spreadsheets is precisely that their processes do not match the standard; forcing processes into the software sends staff back to Excel within three months.

Rebuilding a large system from scratch: a 6-month project, and by the time it is done the processes have changed, staff never tried it, and the spreadsheets still run in parallel "just in case".

The better way is step by step, each step usable immediately.

The 4-stage roadmap

Stage 1: standardize the data inside the spreadsheets (2 to 3 weeks)

No new software yet. What to do:

  • Choose one source of truth per data type: the customer list lives in one sheet, other sheets reference it.
  • Assign identifiers: every customer and product gets a unique code, never relying on names.
  • Fix the column structure, remove the "extra notes" column holding 5 kinds of information.
  • Document the process: who enters what, when, and how it is checked.

This stage is the cheapest and the most valuable: it forces the business to see the data it actually has. Many companies cut errors in half after stage 1 alone.

Stage 2: move the data into a shared database (3 to 4 weeks)

The standardized data moves into a real database with permissions and history. Spreadsheets can still read from it (using built-in connectors), so staff do not have to change habits yet. The difference: several people work at once without overwriting each other, and every change leaves a trail.

Stage 3: entry screens for the real workflows (4 to 6 weeks)

This is where the business web app appears. Not one large system, but one screen per task: receive an order, issue stock, record a receivable. Each screen shows only the fields that task needs, validates on entry (no negative numbers, no duplicate codes), and works on a phone.

Which workflow first? The one most people touch and where errors are most frequent, usually order intake or inventory. Finish one workflow, let staff use it for real for 2 weeks, then move to the next.

Stage 4: automation and reporting (from week 12)

Once data lives in one place and is entered correctly from the start, the following become simple:

  • Weekly reports generated and sent on schedule.
  • Alerts: stock below threshold, receivables overdue, orders unprocessed for 24 hours.
  • Connections to other systems: accounting software, the online store, a Zalo mini app for customer self-service lookups.
  • AI for repetitive work such as reading documents or summarizing commentary; see AI for small business.

Reference costs

  • Stage 1: mostly internal effort, optionally 1 to 2 consulting sessions.
  • Stages 2 and 3: a business web app for 2 to 3 workflows typically runs 40 to 120 million dong depending on complexity and integrations; fixed quotes per scope.
  • Monthly operations: server plus custom software maintenance from 2,000,000 dong a month for a web app, including monitoring, tested backups, updates and change tickets.

Compared with one full-time data entry employee or the damage from a single wrong stock figure, this usually pays back within the first year.

What keeps the system from sliding back to spreadsheets

  • Staff involved from stage 1: the people entering data daily know the exceptions better than anyone.
  • Every stage usable immediately: no "wait 6 more months to see anything".
  • Someone operating it after handover: small bugs fixed the same day via tickets, never piling up into "the software doesn't work".
  • The data belongs to the business: source code, server and database are in the company's name and exportable at any time.

Siri9 does the building under business web app development and the keeping-it-running under software maintenance; both share one service portal so every request leaves a trail.

Frequently asked questions

Does a 10-person business need a business web app?

It depends on how many people touch the data, not headcount. 3 people entering the same inventory file daily is reason enough; 10 people each with their own file is not.

Will we lose old data in the move?

No. Stages 1 and 2 exist precisely to move old data into a clean structure with reconciliation. The old spreadsheets are kept as an archive.

Should we build a web app or buy packaged software?

If 80 percent of your processes match the industry standard, buy and connect the rest. If your processes are your competitive edge, a custom web app preserves that edge. Siri9 advises on both, free of charge.

Where do we start?

Send Siri9 the 2 or 3 spreadsheets you use most (with sensitive data hidden) and a note on who enters what; we reply within 24 hours on which stage to do first and the estimated cost.