Picture an engineer finishing a job, photographing the completed sheet and sending it to the office on WhatsApp. The office manager opens the image, checks the handwriting and types the details into a job record. A missing materials description starts another conversation.
That is the situation I'd want to understand before helping you digitise paper forms. Where does the information go? Who needs it? What stops the job moving forward?
A useful digital job sheet should leave the office with a record it can act on. Here is how I'd approach the change, starting with the work around the form.
Digitise paper forms around the next decision
Take a completed paper sheet and sit beside the person who receives it. Ask them to show you what they do with each answer. Follow the work through to its destination: an invoice, a return visit, a purchase request or a customer report.
Mark the fields they use. Ask about everything left over. Perhaps a box exists because an old customer once requested it. Perhaps nobody knows why the engineer writes the site address when the office already has it against the job.
My recommendation is to give each field a clear purpose before rebuilding it. Keep the information needed to identify the work, explain what happened and decide what comes next. Challenge anything collected simply because there is space on the page.
For an illustrative service visit, I'd consider:
- The job reference and site, selected from the current job list where practical.
- The engineer responsible and the visit date.
- Work completed, with enough detail for someone else to understand it.
- Materials used, linked to the job.
- Outstanding work and the person who should pick it up.
- Supporting photos where they help explain the record.
Treat that as a starting point for discussion. A workshop repair and a contractor's site visit deserve different questions. Leave specialist or regulated records to a separate review with the people responsible for them.
If the destination is unclear, map the handover first. Our guide to swim lane diagrams gives you a way to lay out who does what.
Make the form usable where the work happens
I'd build the smallest useful version and put it on the phone an engineer actually carries. Ask them to complete it beside the equipment, with the sort of interruptions they normally face.
Watch where they pause. Can they find the job? Do the answer options make sense? Is there a sensible way to say access was unavailable? Does the person know whether the record has been submitted?
Use specific questions. Replace a broad “Comments” box with prompts such as “What work remains?” and “What prevented completion?” Give people room to explain an unusual situation without making every routine answer a paragraph.
Where the tool supports it, show follow-up questions only when relevant. Microsoft Forms supports branching based on answers, so respondents can skip questions that do not apply. I'd use that sparingly: a reported defect might open a request for its location and a supporting photo.
Also test the route used by subcontractors. Microsoft says file-upload questions in Forms require an organisation-restricted response setting. Check that restriction before choosing it for a process where external engineers need to upload site photos.
A form that works for its author still needs a proper trial with its intended users.
Test poor signal before choosing the tool
Ask the team where they lose reception. Include those locations in your trial. If someone needs to record work in a plant room, test the proposed approach there before agreeing the rollout.
I'd want to see what happens when the connection disappears during entry, when the user closes the app and when the device reconnects. Check whether the completed record reaches the office and whether the photos remain attached to the correct job.
Be precise about any offline claim. Microsoft's documentation says offline capability for Power Apps canvas apps uses the native Power Apps Mobile players; browser-based canvas apps cannot run offline. Check that product constraint against your working environment before choosing an approach.
For a trial, write down the expected behaviour in plain language: the engineer should be able to see whether a record is saved locally, waiting to send or received by the office. Ask whoever supplies or configures the tool to demonstrate that behaviour on your devices.
Agree a fallback too. If submission fails, the engineer needs a known route for recording the work and flagging the problem. Give someone responsibility for reconciling fallback records when service returns.
Connect submission to a clear office handover
Decide what the office should receive before adding an automation. I'd start with an agreed destination, a named person responsible for checking it and a clear definition of a usable submission.
In an illustrative setup, a completed job sheet enters an office review queue. The office manager checks whether the work is complete, the materials are understandable and any follow-up has an owner. The job then moves to ready for invoicing or back for clarification.
That distinction matters. Pressing Submit should have an agreed meaning. It could mean the engineer has finished entering the visit details, while office review remains outstanding. Make the status labels match those actual steps.
Give corrections a deliberate route. Decide who can amend a submitted record, how a query returns to the engineer and how the reviewer sees that the query has been resolved. Test the awkward cases alongside a straightforward completed job.
I would hold off on automatically creating invoices until that handover is reliable. Get the right information to the right record first. Our article on stopping double data entry covers the wider site-to-office connection.
Choose a tool against the job you need done
Write your requirements from the trial, then compare the options. If you already use a job-management system, ask its supplier to demonstrate your sheet and handover in that system. Include a real awkward case in the demonstration.
A simple online form may be worth testing for a straightforward, connected workflow. Microsoft's Forms creation guide lists choice, text and date questions, alongside other question types. Use those basic capabilities as a starting point only if they fit the process you have mapped.
I would lean towards a more connected job-management approach when the engineer needs existing job details, a history of visits, repeat asset records or a dependable offline workflow. Ask suppliers to show the whole journey from opening the assigned job to office review.
Check the ongoing work as well. Who changes a question when the service changes? Who maintains the job list? Who investigates a submission that has failed to arrive? Include licences, configuration and support in your decision, without assuming an existing subscription covers everything.
The simplest suitable option is the one your team can keep working reliably.
Pilot a complete handover, then retire the old sheet
This week, choose a routine form and agree a bounded trial with an engineer and its office recipient. Keep a note of the current friction: clarification calls, missing details and information entered again elsewhere. Use that note to judge the new approach.
Follow submitted records all the way through office review. Ask whether the recipient could make the next decision from the information provided. Where they could not, change the question, the answer options or the handover.
Before wider rollout, name the owner of the form, agree how problems are reported and decide when the old version will be withdrawn. Keep a fallback available for failures, with a clear route back into the main record.
I would judge the change by whether a job reaches the next person ready to move. That is the standard worth designing for.
If your job sheets keep coming back as office queries, talk to us about improving the handover.