How to prove contractor work was actually done

You know this text thread. "Faucet's fixed 👍," sent Tuesday, read twice, no photo. Three weeks later an invoice arrives for the faucet, plus a shutoff valve you don't remember discussing. You approve it, because what's the alternative? You can't prove anything either way. The unit is forty minutes away and the tenant moved out on Sunday.

Multiply that by every job across every building, and you're running a business on trust plus vibes. Most contractors are honest. That's not the point. The point is that the day someone asks about that repair (a tenant disputing a deposit deduction, an insurance adjuster wanting dates, a buyer's inspector, a judge, or just you six months from now trying to remember whether that water heater part was replaced or repaired), you have nothing. Not because anyone lied. Because nobody wrote it down in a form that survives.

What a real proof-of-work record needs

"Get photos" is the advice everyone gives, and it's half right. A photo on its own proves almost nothing. You probably have hundreds of orphaned repair photos right now that you couldn't match to a job if your deposit dispute depended on it. A record that actually holds up needs four properties:

• Tied to the job and the unit. Not "a photo of a faucet." A photo attached to the job at that unit, alongside the original request and the invoice. Context is what turns a picture into evidence.
• Timestamped by the system, not by memory. "It was done sometime in March" loses to a dated record every time. The timestamp has to be applied when the photo lands, automatically. A date someone types in later is just memory wearing a costume.
• Before and after. The after photo proves the work is done. The before photo proves the work was needed, which is what actually gets contested, in deposit disputes and owner conversations alike.
• Who approved it. Somebody with authority looked at the evidence and accepted the work. That closes the loop: the record moves from "work happened" to "the right person signed off on it, on this date."

Why camera rolls and text threads fail

Most managers already collect some photos. The problem is where they live.

The camera roll is a bucket with no labels. Which unit is this bathroom? Which of the three toilet jobs this quarter is this flange from? You can search a camera roll by date and location, not by job, unit, or building. And half the photos aren't even on your phone; they're on the contractor's, and they leave when he does.

The text thread is worse, because it feels like a record while quietly not being one. It's ordered by time, not by job: the faucet photo sits between a question about a different building and a meme. Group threads scroll fast enough that finding one photo from March means minutes of thumb archaeology. Threads get deleted, phones get replaced, contractors get new numbers. And nothing in a text thread captures the last mile: was this accepted? A thumbs-up emoji is not an approval trail.

Fix the process first

This is a process problem before it's a software problem, and you can start fixing it with the tools you already have. Four rules:

• Photos at completion, before the invoice gets approved. Make it the working rule, not a favor: finished work comes with pictures, and pictures are what get invoices paid. Said once, enforced always.
• One thread per job. Even in email: one subject line per job, everything about that job replies into it. Never let two jobs share a conversation.
• Before and after on anything replaced or repaired. Thirty extra seconds for the contractor. It becomes automatic within a few weeks.
• A named approver. Someone looks at the photos and says "accepted" (in writing, in the thread) before the job is called done.

You can run this on email and folders. Plenty of careful managers do. The catch is that it costs discipline forever (every job, every contractor, every busy week), and the record is only as good as your worst day. The whole system depends on nobody ever taking the shortcut, and the shortcut is always right there.

Then let software hold the line for you

This is exactly the shape RQ200 is built around. Every job carries its building, unit, and tenant contact. The contractor works from a phone view of their jobs, posts photo and video updates as they go, and when the work is finished they submit the job for review, with photos attached, instead of texting "done." You look at the evidence and either approve it or send it back with a note. Only an approval closes the job.

The four properties above stop being rules anyone has to remember. Photos are tied to the job and unit because that's the only place they can go. Timestamps are automatic. The approval is a recorded step, not an emoji. And the record accretes on its own: after a few months you have a dated, photographed, approved history of every repair in the portfolio, built as a side effect of running the work rather than as a chore on top of it.

Will contractors push back on submitting photos? Some will, for a few weeks. That's normal, and it's manageable. We wrote about exactly that in how to get contractors to actually use your software. The short version: phone-first, one new habit, and a manager who holds the line.

The day someone asks about that repair, the answer shouldn't be a shrug and a scroll through old texts. It should be thirty seconds: here's the job, here's the before, here's the after, here's who approved it and when.