A tap washer, a trap and a code nobody picked
Your plumber attends a repair order for a dripping kitchen tap. He replaces the washer, notices the waste trap is leaking, replaces that too, and clears a partial blockage. The job sheet says 'fixed tap and trap'. In the office, someone looks up codes in the schedule of rates, picks the tap one, misses the blockage, and submits.
Across hundreds of repairs a month, the missed items add up, and some claims are rejected for codes that don't match the original order.
Nobody is cutting corners. The operative did good work, and the office did its best with the information it had.
Why coding goes wrong
Schedules of rates are long, specific and differ between landlords. Operatives know the work, not the codes. Office staff know the codes, not what was actually done. The job sheet sits between them with a one-line description.
Variations from the raised order need the landlord's approval under most contracts, and they're often recorded after the fact, if at all.
| Coding step | Common problem |
|---|---|
| Order raised by landlord | Code may not match the real fault |
| Work done | Extra items not recorded |
| Job sheet | One-line description |
| Office coding | Codes picked from memory |
| Submission | Rejected for mismatches |
Operatives also work fast. Repairs contracts are about volume, and a plumber might do several jobs in a morning across an estate. Stopping to look up codes on each one isn't realistic, so the codes are left for later, and later is often the end of the week.
What miscoding costs
Under-claimed work is simply lost revenue. Rejected claims take office time to rework and delay payment. Landlords judge contractors on how clean their submissions are. And without accurate coding you can't see which job types actually make money on the contract. Contract terms are for you and your commercial adviser; we deal with capturing the work accurately.
It also affects performance measures. Landlords track first-time fixes and repeat repairs, and jobs coded wrongly can look like failures or repeats when they are not.
The coding step we build
- Schedule loaded: your contract's schedule of rates is loaded with codes, descriptions and units.
- Work description: the operative describes what they did on their phone, by voice or text, with photos.
- Code suggestions: likely codes are suggested from the schedule, and the operative confirms, changes or removes each. Nothing is submitted on a suggestion alone.
- Variation flag: codes that differ from the raised order are flagged, and a variation request can be sent to the landlord from the job.
- Office check: the office reviews each day's coded jobs, with photos, before submission.
- Submission: coded jobs are sent to the landlord's repairs system in their format, or through their portal or API where available.
Schedules are updated when the landlord issues a new version, and jobs raised before the change keep the codes that applied at the time.
Claims that match the work
Operatives record what they did in their own words. Codes are consistent across the team. Variations are raised on the day. The office checks rather than guesses, and fewer claims come back. Over time, you can see the real mix of work on each contract.
Operatives' voice notes also give the office a clearer picture of what happened in each home, which helps when a tenant complains or a landlord asks why a job needed more than was ordered.
Does your repairs contract work like this?
- SOR codes are picked in the office from job sheet notes.
- Extra work on site isn't coded or claimed.
- Claims are rejected for code mismatches.
- Variations are raised late or not at all.
- You can't tell which repair types make money.