A customer files a complaint about a batch shipped in March. Your quality lead goes to the binder, looks for that week's inspection records and, 40 minutes later, finds a sheet where half the fields are ticked "OK" and the measured dimension could be 12.8 or 12.3. The customer is waiting for an answer, and you don't know whether the fault was yours.
This is a common picture in manufacturers with 10–50 employees. Inspection records exist and get filled in, but nobody uses them until something goes wrong. Then they turn out to be illegible, incomplete or missing.
This post will help you decide whether to move inspection records to tablets, and which of three routes to take: Microsoft Forms, Power Apps or a custom web app. We give costs, timelines and the threshold below which paper is simply cheaper.
What paper really costs
A paper inspection record has three costs that never show up on the printing invoice.
Search time. When a complaint comes in, someone flips through binders. We typically see 20–60 minutes per batch, and half a day of the quality lead's time when the complaint covers several weeks of production.
No trend. On paper, every measurement stands alone. Nobody notices that a dimension on the same machine has been creeping up for three weeks and will be out of tolerance next week. To see that, someone would have to type the numbers into Excel. Nobody does.
Fields filled from memory. The operator fills in the form at the end of the shift, not at the machine. They write what "usually comes out". The record formally exists, but it doesn't tell the truth.
The most common mistake when moving to tablets: copying the paper form one-to-one. If the sheet has 60 fields, an app with 60 fields will be filled in exactly like the sheet — with default values.
Three routes and what they cost
Microsoft Forms (or Google Forms). If you have Microsoft 365, Forms is already included. A form with a dozen or so fields, results flowing into an Excel sheet in SharePoint. Setup takes 2–5 working days, with external cost usually PLN 0–4k (about €0–950). Limits: no validation against the tolerances of a specific part, no link to the order number in your ERP, poor behaviour without WiFi. Good enough for simple yes/no checklists.
Power Apps. A tablet app that pulls the list of orders or parts, shows the tolerances for the selected product and refuses to save when a measurement is missing. Data goes to a SharePoint list or Dataverse. Implementation typically takes 3–6 weeks and PLN 15–40k (about €3,500–9,400). Plus licences: Power Apps Premium costs USD 20 per user per month today and USD 22 from 1 January 2027 (source: Microsoft, Power Apps Premium price update). For 8 inspectors that is roughly PLN 750 (about €175) a month. We cover where the platform fits in more detail in our post on Power Platform for SMEs.
Custom web app (e.g. .NET). Runs in the tablet's browser, reads orders and product master data directly from the ERP (in Poland typically Comarch Optima, enova365 or Subiekt nexo), stores measurements in its own database, works offline and syncs when back in range. Typically PLN 40–90k (about €9,400–21,000) and 6–12 weeks, plus 10–15% of that per year for maintenance. No per-user licence fees, but every form change is developer work.
The rule we use: Forms up to 20–30 records a week with no numeric measurements; Power Apps when there are measurements and tolerances but no live ERP link is needed; a custom app when the record must be tied to the order and batch in the ERP, or when the shop floor has no reliable WiFi. We describe the general logic of this decision in our post on build vs buy.
Two examples from our projects
When it worked. One of our clients — a 22-person CNC machining shop supplying the machinery industry — kept in-process inspection records on paper, about 150 a week. Their ERP is Comarch Optima, a common Polish ERP. We built a .NET web app: the inspector scans the order barcode on the routing card, the app pulls the part and its inspection dimensions from Optima, the inspector enters 6–12 measurements, and any value out of tolerance turns red immediately and requires a comment. The project took 7 weeks and cost PLN 54k (about €12,700), plus 4 tablets in rugged cases at about PLN 3k (about €700) each.
Result after six months: finding the records for a batch went from about 40 minutes to under a minute. Responding to a customer complaint went from 2–3 days to the same day. The quality lead gets a weekly chart of measurements per machine and twice caught tool wear before parts went out of tolerance. The lesson: we cut the form from 34 fields on paper to 9 mandatory ones. If we did it again, we would make that cut before the first line of code, not in week three.
When it went wrong. At another company — a 35-person corrugated packaging manufacturer — an employee from the quality department built a Power Apps app in-house. They copied the paper form one-to-one: 58 fields. Nobody checked WiFi coverage in the die-cutting area. Operators filled in the record at the end of the shift next to the office, picking default values. After 4 months half the lines were back on paper. Cost: about 120 hours of internal work, PLN 18k (about €4,200) of external support, plus licences. The second attempt took 3 weeks: form cut to 14 fields, two extra WiFi access points for about PLN 2.5k (about €600), offline mode. The lesson: the problem was not Power Apps, it was the form and the shop floor.
When it doesn't make sense
A tablet does not always beat a sheet of paper. Stay on paper if:
- You have fewer than about 20 records a week and complaints come once a quarter. The implementation cost will not pay back in saved search time.
- Nobody will look at the data. If there is no one who checks the trends once a week, a digital record is just a more expensive binder.
- The shop floor has no coverage and you don't want to invest in it. An app without offline mode in an area without WiFi will be abandoned within a month.
- Your customer or auditor requires a signed paper record. Before you start, ask your largest customer and your certification body whether they accept an electronic record with the person's identity and a timestamp.
What can go wrong even in a good project: tablets on the shop floor break and get dirty. Budget PLN 2.5–5k (about €600–1,200) per device in a rugged case, plus one spare. Work gloves don't work with every touchscreen — test that on the floor before you buy four units. And one more thing: for the first 2–3 weeks operators will be slower than on paper. That is normal, but someone has to be there to keep an eye on it.
What you can do this week
None of these steps requires a purchase or a vendor.
- Count the inspection records from the last 4 weeks and divide by 4. That is your weekly volume.
- Take one paper record and mark the fields someone actually used during the last complaint. The rest are candidates for removal.
- Time how long it takes to find the records for one batch from 3 months ago.
- Walk the shop floor with your phone and check WiFi coverage at every inspection station.
- Ask your largest customer whether they accept electronic inspection records.
After these five steps you know whether you have a problem, how big it is, and which of the three routes fits it.
What next
Digitising inspection records is usually not a six-month project. With a clear form and a checked shop floor, you can launch Forms in a week, Power Apps in a month and a custom app in two to three months. But none of these routes will fix a 60-field form that nobody reads anyway.
A tablet won't improve your quality control. A shorter form filled in at the machine will — the tablet just means you find the result in a minute instead of an hour.