During our last real outage, I watched a customer screenshot our status page next to their own error dashboards. The page said "Investigating." Their dashboard showed a full outage. That screenshot circulated in their Slack, and we spent a quarter rebuilding a relationship that one honest update could have saved.
The failure wasn't dishonesty — it was template language. "We are aware of an issue and are investigating" is technically true and completely useless. Customers don't need confidence; they need predictions.
The three lines that work
Line 1 — What's broken, in the customer's world. "Checkouts are failing for roughly half of attempts." Not "API errors" — say what they see.
Line 2 — What we know and don't. "Cause is in our payment provider integration; card data is unaffected. We don't yet know full scope." Admitting the unknowns is what makes the known parts believable.
Line 3 — Next update at a specific time. "Next update by 14:15 or when anything changes, whichever comes first." Then hit it — even if the update is "no change, still working it." The promise-keeping matters more than the news.
The counterintuitive part
More frequent updates don't rebuild trust — kept promises do. One update every 30 minutes that says exactly what it will say beats five vague ones. When we trained on-call on this format, our "status page screenshot in a complaint" incidents went to zero in two quarters. Customers don't remember that you had an outage. They remember whether you told them the truth about it, on time, in their language.
Every drill in this post ships in the Ops Starter Kit ($29). All five kits in one download: Ops Mega Bundle ($49). Or let the always-on ops desk run them for you.
Top comments (0)