This is a really good point. Getting flagged as "difficult" is a genuine risk I hadn't considered.
Following your approach of stating the SLA breach publicly feels like it walks that line. It's a fact, not an emotion. My worry is that even a factual breach notice could still trigger that internal flag on your account. Is there a point where you're better off just having legal send the notice directly, so the support and account teams can still play the helpful role internally?
Ugh, that's rough. Five days with just an auto-reply is completely unacceptable for a broken financial workflow.
To answer your questions:
1. **Widespread bug?** Hard to say for sure, but the pattern you're describing (approve button works, audit log updates, but state doesn't transition) is exactly the kind of thing that points to a systemic backend regression, not just your instance. A check of their status page (if they have one) or even Twitter/X for "Approve.com workflow" might give clues.
2. **Workaround?** The only ugly workaround I've seen for this specific state-stuck issue is to have your department head *reject* the request with a comment like "SYSTEM: Manual push to finance," and then have the requester immediately submit a new, identical request. It's awful, but it sometimes bypasses the stuck routing engine. You'd need the finance controller to fast-track the "new" request.
3. **Other channels?** Tag them on LinkedIn or X. Seriously. A public, factual post stating "Day 5, no response on critical financial approval outage, ticket #XXXXX" often gets a different team's attention. Also, dig for an account manager or customer success contact in your past emails - they can light a fire internally.
The escalation path also being dead is a huge red flag. That suggests the entire rule engine is offline or broken, not just one step. Good luck, and keep us posted.
Clean code is not an option, it's a sanity measure.
Totally agree on the public channel tactic. A concise, factual post tagging their main account can cut through the noise, especially if it mentions the specific SLA clause they're violating.
One caveat on that workaround though, the "reject and resubmit" method can screw up your audit trail for compliance. You'll lose the original submission timestamp and the full history, which might be a problem later. If you have to do it, screenshot everything before you reject.