What does moving from Excel to a web app involve?
You may run something important in a spreadsheet: the job tracker, the inventory count, the commission sheet, or the weekly numbers the owner reads on Monday. It began as one person’s file and became the place the team works. Now four people edit it, two keep private copies, and putting the Monday numbers together takes half a morning.
Moving from Excel to a web app takes that workbook apart into three layers:
- The records: each row becomes a record in a database, with fields that accept only the right kind of value: a date where a date belongs, a customer picked from a list instead of typed.
- The logic: the formulas become calculations in code, written once, tested, and run the same way for everyone.
- The screens: forms and lists in a browser, each showing a person what their role needs and letting them change only what they should.
A dashboard sits on top of the same records and shows the totals, so nobody rebuilds them by hand. Because it reads the records directly, the figures are as current as the last entry.
What are the limits of managing a business in spreadsheets?
The trouble starts when a workbook becomes the system a team runs on, because nothing in a workbook stops a wrong entry or a broken formula unless someone sets it up to:
- Errors are common and hard to see: Raymond Panko’s 2000 review found that fifteen years of studies had all concluded spreadsheet errors are common and non-trivial. Field audits found errors in 24% of the 367 spreadsheets checked, though most of the older audits used methods unlikely to catch most errors; the most recent audits, with better methods, found errors in at least 86%.
- Errors that change the numbers: in a study of 50 working spreadsheets published in 2009 by Stephen Powell, Kenneth Baker and Barry Lawson, 43 had at least one error that gave a wrong result, 281 instances in all. The most common problem was hard-coding, a number typed into a formula instead of kept in a cell of its own: 37.7% of the 483 error instances they logged, most of which hadn’t produced a wrong result.
- Checking doesn’t catch everything: in Panko’s review, one person going through a spreadsheet cell by cell found half or fewer of its errors, and group inspection caught about 80%.
- Shared editing has conditions: co-authoring needs the workbook on OneDrive, OneDrive for Business or a SharePoint Online library, and Microsoft says the most common cause of a “locked” file is someone opening it in a version of Excel that doesn’t support co-authoring, which locks it for everyone else.
- Size has a ceiling: Microsoft’s published limits, read in November 2026, are 1,048,576 rows by 16,384 columns per worksheet and 32,767 characters per cell.
When should I move from Excel to a custom app?
Move when the spreadsheet has stopped being a place to think and become the system the work runs through. The signs:
- Private copies: files named “final v3” on people’s desktops, because the shared one is locked, slow or easy to break.
- Figures typed twice: totals from the workbook retyped into the accounting tool, the CRM or an email to a client.
- People outside the file need it: a customer wants the status of an order, or a crew needs today’s jobs on a phone.
- Rules that matter: approvals, limits on who may change what, and a need to know who changed it.
- A file that strains: slow to open, slow to recalculate, or split into several workbooks to keep it working.
There are four routes off a spreadsheet, from lightest to heaviest:
| Route | What it is | Fits when |
|---|---|---|
| A tidier shared workbook | One file on OneDrive or SharePoint, co-authored, with tables and protected sheets | A few people, simple rules, and the numbers stay inside the team |
| A low-code app | Power Apps or a similar tool, on an Excel table or a database such as Dataverse | Forms and lists for a team already on Microsoft 365, at modest volume |
| A BI tool over the data | Power BI or a similar tool, reading the workbook or a database for reports | Data entry works, and the trouble is reporting |
| A custom web app | A database, tested logic and screens made for the business | Its own rules, outside users, higher volume, or several systems tied together |
The routes also combine: a BI tool can report on a custom app’s database. Our guide to no-code and low-code tools compares the middle routes in more depth, and build vs buy internal tools covers when an internal app is worth building.
How do I turn an Excel spreadsheet into a web app?
The move goes in a set order, whatever the tool:
- Map the workbook: list each sheet, what it holds, who edits it, and which cells are typed and which are calculated.
- Separate records from reports: sheets of rows (jobs, orders, items) become tables, and sheets of totals become dashboard screens.
- Clean as you move: one spelling per customer, one format per date, and duplicates merged. Numbers hard-coded into formulas become named settings someone can see and change.
- Rebuild the formulas as tested logic: check each calculation against the spreadsheet’s own answers on real past data before you trust it.
- Build the screens by role: the person entering a job sees a form, and the manager sees the list and the totals.
- Run both for a while: keep the old workbook read-only beside the new app until the figures match.
On Microsoft 365, Power Apps can build a canvas app from Excel data, either by uploading the workbook into a Dataverse table or by connecting to a file that stays in cloud storage such as OneDrive, Dropbox or Google Drive. Either way, Microsoft says to format the data as a table first, and Power Apps can connect only to Excel files kept in the cloud (Microsoft Learn, May 2026). Volume is the next limit: by default, Power Apps retrieves only the first 500 records from a data source, and app settings can raise that to 10,000. For larger data sets, Microsoft points to functions that can be delegated to the data source, such as Filter and Sort, or to Dataverse, which supports delegation more broadly (Microsoft Learn, October 2026).
What is a KPI dashboard?
A KPI dashboard is one screen of a business’s key performance indicators, each shown against a goal, so the people reading it can see at a glance what’s on track and what isn’t. Microsoft’s Power BI documentation describes a dashboard as a single page of visuals that tells a story (December 2025), and a KPI as a visual cue that shows progress toward a measurable goal. Its KPI visual needs three things: a base measure, a target, and a threshold or goal (August 2026).
Dashboards and reports do different jobs. In Power BI, a report can run to several pages and lets the reader filter and slice the data, while a dashboard is one page that can’t be filtered but can send an email alert when a set condition is met. The working difference: the dashboard answers “are we on track?”, and the report answers “why not?”
What should a KPI dashboard show?
Start from the decisions. Microsoft’s design tips for Power BI dashboards (October 2025) begin with the audience: how it uses the dashboard, and which key metrics help it make decisions. The same guidance gives a short list of rules:
- One screen: keep every tile on one screen, with no scroll bars, and cut anything that isn’t essential.
- The top-left corner: the highest-level figure goes there, with more detail following in the direction people read.
- Short numbers: three or four numerals at most, so “3.4 million” rather than “3,400,000”.
- Consistent time frames: keep each chart’s period clear and the periods consistent, so last month’s figures don’t sit beside a chart for some other month.
- Plain charts: bars and columns to compare values, and no 3D charts.
Then choose the measures by role. An owner might watch cash, revenue against target and margin by service. A dispatcher watches jobs scheduled, jobs running late and crews available. A sales lead watches new inquiries by source and quotes waiting on an answer. Every figure on the screen should have someone who acts when it moves; a tile nobody acts on is noise.
What is the best KPI dashboard software?
The right tool depends on where your data lives, who needs to see it, and what each viewer costs. The options fall into three groups:
- Reports inside the tools you use: accounting, CRM and job-management software usually has reports and dashboards of its own. Start there when the numbers live in one system.
- A BI tool: business intelligence software such as Power BI connects to several sources and builds dashboards across them. Check the cost per viewer: in Power BI, a colleague needs a Pro or Premium Per User license to view a dashboard you share, unless the content comes from a Premium capacity or a Fabric capacity of F64 or larger (Microsoft Learn, December 2025).
- Custom business intelligence software: dashboards built into a system made for your business, reading its database directly, with the same logins and permissions as the rest of the app. It suits figures that depend on your own rules, and dashboards for customers or crews who would otherwise each need a license.
Before you choose, a business intelligence consultation, with a vendor, an outside consultant or your own team, should settle four things: which decisions each dashboard serves, where each figure comes from, who owns each number, and how fresh it has to be.
What goes wrong when a business moves off a spreadsheet?
Moves off a spreadsheet tend to go wrong in the same few places:
- Copying the mess: every sheet, column and workaround rebuilt in the new app. Use the move to drop what nobody uses.
- Skipping the cleanup: duplicates and inconsistent names carry into the database, and the dashboard reports them faithfully.
- No parallel run: switching over on a Monday without checking the new totals against the old ones.
- Leaving out the keeper: the person who looked after the spreadsheet knows why each odd formula exists. Build it with them.
- A dashboard nobody opens: forty tiles picked by whoever built it. Let the people who read it pick the five they need.
- No way out: check that you can export your records in a standard format from whatever you choose.
Where do you start moving from Excel to a web app?
- Name the workbook the business runs on: the one people would miss most if it disappeared.
- List who touches it and why: each role, what they enter and what they read.
- Mark where it hurts: figures retyped, copies made, errors that have cost money.
- Pick the route: a tidier workbook, a low-code app, a BI tool or a custom app, from the table above.
- Clean one table first: the customers or the jobs, so the first screen runs on good data.
- Run old and new side by side: switch over once the totals have matched for a few weeks.
If the workbook has outgrown itself, our dashboards and reporting work moves the records into a database and builds the screens and dashboards on top, in a system made for your business alone and kept in a repository of its own, and once it’s live we run it on a monthly plan. If a tidier workbook or a tool you can subscribe to would cover it, we’ll say so on a 15-minute call.
Sources
9 cited
- Microsoft Support, Excel specifications and limits, read 2026-11-01
- Microsoft Support, Collaborate on Excel workbooks at the same time with co-authoring, read 2026-11-01
- Microsoft Learn, Create a canvas app based on Excel data, last updated 2026-05-19, read 2026-11-01
- Microsoft Learn, Understand data sources for canvas apps, last updated 2026-10-07, read 2026-11-01
- Microsoft Learn, Dashboards for business users of the Power BI service, last updated 2025-12-01, read 2026-11-01
- Microsoft Learn, Key performance indicator (KPI) visuals, last updated 2026-08-24, read 2026-11-01
- Microsoft Learn, Tips for designing a great Power BI dashboard, last updated 2025-10-01, read 2026-11-01
- Raymond R. Panko, Spreadsheet Errors: What We Know. What We Think We Can Do, EuSpRIG 2000 proceedings, arXiv, read 2026-11-01
- Stephen G. Powell, Kenneth R. Baker and Barry Lawson, Errors in Operational Spreadsheets, Journal of Organizational and End User Computing 21(3), July-September 2009, read 2026-11-01