Communicating with Customers During a Disruption: What to Say Now vs Later
When systems fail, the words a company chooses can either rebuild trust or destroy it permanently. This guide draws on expert recommendations to show exactly what to communicate in the critical first hours of a disruption and what information to save for later. The three strategies outlined here help organizations respond with clarity and confidence when customers need answers most.
Announce Failure, Promise Cadence, Avoid ETAs
Hi, I'm reaching out from a PR agency to share a SaaS founder's approach to communicating with users during a service disruption.
- Kevin Lourd, Founder
- distribute (https://distribute.you)
- Photo URL: https://media.licdn.com/dms/image/v2/D5603AQEVewo3v561Qg/profile-displayphoto-crop_800_800/B56Z1I_iAFJYAI-/0/1775046110821?e=1781740800&v=beta&t=SthaA3wMf_28mNQhspliRTI6ZB7XbIsUaSlPb3wGQTw
- LinkedIn: https://www.linkedin.com/in/kevin-lourd-3394b025/
- Bio: Founder of distribute, a single dashboard for builders to automate outbound distribution across sales, PR, VCs, hiring, and accelerators using AI.
Here's Kevin's answer:
"My company builds software that automates outbound sales and PR for startup founders. When our system goes down, our users are actively missing potential deals by the minute, so they panic easily. During a disruption, I decide what to say by drawing a hard line between what is broken and when it will be fixed. I will immediately broadcast exactly what is failing, but I usually hold back any timeline for a resolution until our engineers actually deploy a patch. A while back, one of our critical data providers started dropping connections, pausing all our users' active outbound campaigns. Instead of sending an optimistic 'we are working on a fix' email, our timing choice was to commit to a rigid update cadence. We sent an alert saying: 'Our primary data provider is currently failing, which means your active campaigns are paused. No data is lost, but nothing is sending. We will update this status page every 30 minutes until we are back online.' We didn't promise an ETA. We just kept updating that page exactly on the hour and half-hour, even if the update was just 'Still down, investigating.' It calmed our users because they didn't have to wonder if we had forgotten them, and when we finally routed traffic through a backup provider a few hours later, we hadn't broken any premature promises about when it would be resolved."
Show Containment First, Explain Causes Later
During a disruption we first answer whether the issue is contained or spreading in the system. We focus on clear boundaries instead of detailed explanations in the first message to customers. We explain what is affected what is not affected and how we are stopping further drift while we investigate. This keeps the message useful without assuming full certainty at that stage.
One timing choice that worked well was holding corrective detail until containment was in place. We share protective action first and share diagnosis later once stability is reached. Customers stay calmer because the update shows control before deeper analysis begins. This approach builds trust since we demonstrate stability first and explain causes after clarity improves gradually.
State Facts Now, Set Next Check‑In
I run a specialist online retailer of EV charging cables, so when something breaks for us it is usually orders stuck with a courier, a payment glitch at checkout, or stock that has gone missing between the supplier and our shelves. The instinct under pressure is either to go silent until you have the full picture or to promise a fix you cannot yet stand behind. Both lose trust. Silence reads as not caring, and a confident promise that later unravels is worse than the original problem.
The line I hold is to say the true thing I already know now, and explicitly hold the thing I do not. So I will tell a customer straight away that their parcel is delayed and that I am chasing the carrier, but I will not invent a delivery date I cannot guarantee. What I commit to instead is a next update by a stated time, even if that update is only that I still do not have an answer. People can sit with a problem far better than they can sit with being ignored, and a promised check-in that I keep does more for trust than any reassurance.
The choice that has held up best was during a courier backlog that hit a batch of our orders at once. Rather than wait for the carrier to sort itself out, I emailed every affected customer the same morning, told them plainly what had happened, said I did not yet know exactly when each parcel would move, and gave them a firm point to expect the next message. Almost nobody asked for a refund. Roughly 85% simply replied to say thanks for the heads-up, which is not a number I would have got by going quiet and hoping.
The thing that survives later scrutiny is the update you held back, not the one you rushed. Tell people what is real, name what you are still finding out, and never let a fixed promise outrun what you know for certain.
Lead With Safety Steps, Defer Findings
Start with clear safety steps that protect people and data right now. Say what to stop doing, what to keep doing, and which tools are safe to use. State that the cause is under review and will be shared after facts are confirmed.
Give a time for the next update so customers know when to check back. Keep the tone calm and direct to lower stress. Publish a safety‑first notice now and schedule a fact‑based cause report for later, then act today.
Provide Workarounds, Note Limits, Schedule Update
Offer simple workaround steps that help customers keep working during the issue. Keep the guidance short, clear, and tied to common tasks. Do not guess about why the problem exists or how long it will last.
Promise a next update time and stick to it. Note any limits or risks of the workaround so users can decide. Share the workaround steps now and hold back any theories until the investigation is done, then act now.
Direct Customers To Live Support, Reserve Details
Make sure customers know exactly how to reach help right away. List the active support channels, expected wait times, and hours of coverage in plain words. Provide an incident number so every contact lines up to the same case.
Avoid sending raw diagnostics or logs until they are reviewed and safe to share. Tell customers when deeper findings will be shared and by whom. Point customers to the live support channel now and reserve detailed diagnostics for a verified update, then reach out today.
Summarize Symptoms, Confirm Sources, Verify Fixes Afterward
Describe what users can see and measure, not what might be causing it. Name the affected features, the time the issue started, and any error messages on screen. Share who is impacted by region or account type if known.
Avoid guesses about sources, attackers, or vendors until proof is solid. Promise a follow‑up that explains confirmed causes and fixes. Post a clear symptom summary now and commit to cause details only after they are verified, then communicate now.
Dispatch Compliant Notices, Document Everything, Delay Attribution
Send required legal or regulatory notices within the mandated time window. Use approved language that covers scope, impact, and steps taken, and keep it consistent across channels. Do not assign blame to third parties or name threat actors until evidence is confirmed and counsel agrees.
Keep records of what was sent, when it was sent, and to whom it was sent. Tell customers that a full attribution and timeline will follow after the formal review. Dispatch the compliant notice now and save any attribution for a validated post‑incident report, then start today.