How do you handle non-technical “system outages” with customers?
Uh oh
Have you ever received an email from an upset customer or partner? Speaking from experience, this kind of situation causes the same feeling of dread as a production outage and deserves similar treatment. Anytime our EPX (Engineering Partner Experience) team lets a startup partner down, we follow specific steps to remedy the situation. A startup partner wrote us last week, upset and concerned over a problem. Here’s how we respond to these issues: apologize, take ownership, overcorrect.
- Apologize
In each of these situations, we respond immediately and make a real apology. The first person who becomes aware of the problem reacts and responds as soon as possible — that day and no later than the next day.
2. Take Ownership
All relevant stakeholders get together that day or the following day to conduct a blameless post-mortem. We take an in-depth look at what happened, when, how we responded, and how we can bulletproof ourselves for the future. “Blameless” is critical because we see these situations as a source of learning and data, not something to bring on feelings of shame.
3. Overcorrect
Within 24 hours, we send the affected customer or partner an executive summary of the situation, including details on all the areas we examined. For our learning, we log each of these system outages in a central location. Ideally, those action items have a timeline or a due date to follow up when we have a resolution.
Do you do this too? I’m curious what kind of framework you use for non-technical “system outages.”
