Open research · July 2026
What 7,682 field service complaints actually say
Including the hypothesis we started with, and had to throw away.
Method
How this was measured
The app universe was built by sweeping 30 trade-specific queries against the iTunes Search API — “hvac business software”, “work order management”, “pest control software”, and so on. That returned 495 apps, filtered down to 49 genuine field service tools; job boards, garage-door openers, retail shift-schedulers and consumer lawn apps were removed. Google Play packages were resolved by verifying candidate package names against the store's own detail endpoint, because Play's search silently returns wrong apps.
Reviews rated one to three stars were collected from the Apple App Store across four English storefronts and from Google Play, deduplicated by review ID. Each review was tagged against a 23-theme complaint taxonomy using pattern matching on its title and body. Theme percentages are computed only over substantive reviews — those of 60 characters or more — because a review reading “Terrible” carries a rating but no diagnosable complaint and would otherwise deflate every rate.
Finding 01
Reliability is the whole story
Crashes, freezes and “it doesn't work” account for 19.5% of substantive complaints — more than missing features, and nearly three times the rate of complaints about the software being hard to learn.
| Complaint theme | Share |
|---|---|
| Reliability — crash, freeze, doesn't work | 19.5% |
| Missing feature | 11.2% |
| Billing & cancellation | 8.3% |
| Customer support | 7.0% |
| Complexity / onboarding | 6.8% |
| Price & value | 6.3% |
| Team & permissions | 5.8% |
| Performance / slowness | 5.0% |
| Photo & attachment handling | 3.6% |
| Offline & connectivity | 1.7% |
It is also universal. This is not one bad vendor dragging an average — every app measured with 100 or more substantive low-star reviews falls between 11% and 41%.
| App | Reliability complaints | Reviews |
|---|---|---|
| Salesforce Field Service | 41.0% | 217 |
| Workiz | 34.2% | 155 |
| ServiceTitan Mobile | 31.9% | 596 |
| ServiceTitan Field | 27.4% | 186 |
| Simpro Mobile | 26.1% | 111 |
| Buildertrend | 20.3% | 349 |
| Jobber | 19.5% | 472 |
| Housecall Pro | 13.9% | 410 |
| Joist | 11.1% | 334 |
Finding 02
The hypothesis we abandoned
This study began with a different conclusion. An early pass over six apps found 29 complaints about the app failing without signal, and produced a confident thesis: field service software dies where the work happens, and offline-first is the opening.
Widening the universe from 6 apps to 21 dropped the offline share to 2.4%. Widening it again to 58 apps dropped it to 1.7%. A signal that is genuinely market-wide holds its share as the sample grows. One that shrinks was an artefact of which apps you happened to look at — in this case, Jobber and Housecall Pro, which do have unusually high offline complaint rates (3.1% and 3.3%) relative to everyone else.
We are publishing this because it is the most useful thing in the study. The offline thesis was plausible, well-evidenced at small scale, and wrong.
Finding 03
Photos are the work product
Photo and attachment handling draws 3.6% of complaints — more than double the offline signal — and when connectivity does fail, photos are what breaks most (23.5% of those cases), ahead of invoicing and payment collection (20.2%).
“Needs an option to upload multiple photos from gallery at once, instead of one at a time. Needs ability to remove photos from tasks in case of accidental upload.”
UpKeep · 1★ · 54 votes
“Often times I'll attempt to upload pictures and it won't do it, OR it'll say it won't upload when it actually did and I end up uploading duplicates.”
Buildertrend · 1★ · 17 votes
That second quote is the offline problem correctly scoped. The failure is not the absence of an offline mode — it is an upload queue that reports a state it isn't in.
Finding 04
Owners complain about money; technicians complain about crashes
Complaints about money — billing, cancellation, and price — total 14.6%, the second largest cluster. They rose sharply when Apple data entered the corpus, which is also where owner-voice reviews concentrate. Crash complaints run the other way, skewing toward the people who were handed the tool rather than the people who bought it.
“As a technician this app is, on average, the worst part of my day.”
Jobber · 1★ · 24 votes
The commercial implication is uncomfortable and worth stating plainly: reliability is what the daily user suffers, but money and support are what the buyer talks about. A reliability-first product is a retention argument more than a sales argument.
Limitations
What this study cannot tell you
- 41% of substantive reviews were not matched by the taxonomy — idiosyncratic one-off narratives. The miss rate is roughly uniform across apps, so comparisons hold, but every percentage here is a floor rather than a true rate.
- App store reviews over-represent the daily user. Buyers complain on G2, Capterra and in churn calls. Those sources block automated collection and are not represented here.
- The technician-versus-owner split is inferred, not measured. An attempt to classify reviews by speaker left 87.6% unattributable by pattern matching. Finding 04 rests on the shift in theme mix between stores plus individual quotes — it is directional, not proven.
- No recency weighting. A complaint from 2021 counts the same as one from last week; some cited failures may already be fixed.
- Two apps carry outsized weight. ServiceTitan Mobile (596) and Jobber (472) contribute the most substantive reviews, so the corpus leans toward larger vendors.
We're building against this data
Crewproof is a field service app for small crews, built reliability-first. In development — the waitlist is on the home page.