Why telling customers first changes how a problem lands
Most service problems are discovered by customers in one of three ways: they wait for something that does not happen, they try to use something that does not work, or they hear about it from someone else. Each of those moments produces a contact, and each contact starts with a customer who already feels let down and has had to do the work of finding out. The same problem, announced by the business with a clear explanation and a choice, starts in a different place: the customer has been told, has options and knows when they will hear more.
The delivery promises guide puts the rule plainly for broken delivery dates: tell affected customers before they ask, with the new date, a one-sentence reason and what they can do if the new date no longer works for them, and keep the delay notice a service message. That guide owns delivery. This guide applies the same habit to any operational problem in any business: a fault in a service, appointments that have to be cancelled, an event or trip that cannot run as booked, planned maintenance that will interrupt something customers rely on.
Proactive messages are part of recovery rather than a substitute for it. The service recovery and complaints guide sets out the recovery sequence, from acknowledging the problem to following up and learning from it. A proactive message is the acknowledgement, sent before the customer has to ask for it. The fix, any amends and the follow-up still have to happen.
This guide stops short of a crisis. When the whole business is in trouble, the guide to leading marketing through a business crisis covers pausing promotion and keeping every message in line with the business's statement, and the social media crisis plan covers the first hour in public, with holding statementsA brief first public response in a crisis that acknowledges the issue, tells affected people what to do now and says when more will follow. drafted in advance. Planned changes with notice, such as retiring a feature or communicating a price change, have their own guides too. When planned bad news is likely to reach journalists, the guide to announcing bad news to the press sets where the press comes in the order of telling. What is left is the everyday disruption: real, contained and affecting a group of customers the business can name.
When to send and who decides
Decide in advance who can send a disruption message, because the decision is usually needed when the person who normally approves messages is busy with the problem itself. Name an owner for operational messages, often the head of operations or support, with a deputy. Give them a short rule for when to send: the problem will change something a customer was promised or is relying on, the business knows enough to say what changes for them, and the customer would want to know now rather than later.
The second condition matters. A message sent before anyone understands the problem tends to be vague or wrong and needs correcting. Waiting for perfect information is the opposite mistake: customers start to notice and contact the business while it is still drafting. The workable middle is to send once the business can state what has happened in one sentence, who is affected, what changes for them and when they will hear next, even if the cause and the fix are not yet known.
Some problems do not need a message. A fault fixed before it reaches any customer, or a delay absorbed inside the original promise, changes nothing the customer was told. Sending anyway teaches customers to ignore service messages.
When the problem escalates beyond the operational, such as a safety issue, a data problem, or something attracting public attention, the decision moves to whoever owns crises and legal questions in the business. The operational owner should know where that line sits before it is reached.
Finding who is affected
The message goes to the customers the problem affects, and only to them. Sending a disruption notice to everyone alarms customers who were never affected and generates contacts from them; sending it to too few leaves some affected customers to find out the hard way.
Start from the system that holds the commitment: the order system for deliveries, the booking system for appointments and trips, the account records for a service. Define the affected group by the facts of the problem, such as orders dispatched from one warehouse on certain days, bookings on a cancelled route, or accounts on a service in one area. Check the list against a second source where you can, because a filter that is slightly wrong is the most common reason a disruption message reaches the wrong people. Note customers who need a different message, such as those with something time-critical riding on the booking, and customers already in contact with support about it.
Keep the list. It is needed for the updates, the closing message and any remedy, and it lets the business measure later how many of the affected customers still had to get in touch.
What the message says: a table
Write the message on the inverted pyramid: what it means for the reader first, then what they can do, then the background. A customer who reads only the subject line and the first sentence should know whether they are affected and what changes. The message is a Transactional EmailAn email triggered by a person's own action or account, such as a receipt, password reset or shipping notice, rather than a marketing send. or its equivalent in other channels, and it is read as one of the most important service touchpointsAny interaction between a person and a brand, paid, owned or earned, before, during or after a purchase. the business has with that customer.
| Part | What it says | Example wording |
|---|---|---|
| Subject or first line | That something has changed for this customer, in plain words | "Your Tuesday appointment has had to be cancelled" |
| What changes for them | The specific effect on their order, booking or service | "Your engineer visit on Tuesday morning will not go ahead" |
| Their choices | What they can do now, with how to do it | "Choose a new slot using the link below, or reply and we will call you" |
| The reason | One honest sentence, without blame or jargon | "Our engineer for your area is unwell and we cannot cover the visit" |
| The apology | Specific, for the failure, owned by the business | "We are sorry; we know you arranged to be in" |
| What happens next | The next update and when it will come, or that this is the final word | "We will update you by Monday afternoon, whether or not anything changes" |
| Where to get help | The help page and the contact route, with any extended hours | "Questions: reply to this email or call the number below" |
| Nothing else | No offers, cross-sells or newsletter content |
The last row matters. The UK regulator's direct marketing guidance, as the guide to promotional and transactional email describes it, separates service messages a customer needs from marketing a business wants to send, and treats the content and purpose of a message as what decides which it is. A service message that carries a promotion starts to look like marketing, to recipients and to regulators. A disruption notice therefore stays free of offers. If the business wants to recognise the trouble it has caused, that is a remedy for the affected customers, set and explained as such; the tutorial on setting remedies and goodwill limits sets out how remedies are decided. Rules vary by country, and this is general direction rather than legal advice. When one failure has affected a whole group, the guide to deciding a remedy when one failure affects many customers covers who qualifies, whether the remedy is applied without a claim and how this message explains it.
Channels, the help centre and the status line
Use the channel the customer gave for service messages about this order, booking or account, usually email, and add a faster channel when the change is urgent and the customer has agreed to receive service messages there, such as a text message for a cancellation the next morning. A message the customer will not see until after the event is not proactive. The support channels and self-service guide covers which channels suit which questions; for disruption, speed and reach decide.
Then make the same facts available where customers who did not see the message will look. Put a short notice on the relevant help article or on a page set aside for current problems, and a line on the contact page or in the phone greeting, so customers who call are told before they wait. Some businesses call this a status line: a single place, dated and updated, that says what is affected, what customers should do and when it will next be updated. Customers who were not affected can check it and move on.
Social channels follow the same facts. If customers are likely to ask in public, a short post or a pinned reply with the same wording saves dozens of individual replies, and the social team should have the message before it goes out. If the original message went out with a mistake in it, such as the wrong date or the wrong group of customers, the tutorial on fixing a mistake in a sent email covers the correction.
Updates, the support team and closing the problem
Every message promises the next one. State when the next update will come, and send it then, even if it says the position is unchanged. Silence after a promise of an update undoes the trust the first message earned. Updates repeat the essentials for anyone reading only the latest message: what is affected, what the customer can do, when they will hear next. Update the status line at the same time.
Brief the support team before the message goes out, not after the first calls arrive. Agents need the same facts the customer has, the affected list or a way to check whether a caller is on it, the choices on offer, what they can give without asking, what to escalate, and the time of the next update. The habit is the same one the tutorial on briefing customer service before a campaign builds for planned marketing, applied to a problem nobody planned. A support team hearing about a disruption from customers is the clearest sign the process failed.
Close the problem with a final message to everyone who received the first one: the problem is resolved, what the customer needs to do now, if anything, and where to go if something is still wrong. Remove or date the status line notice. Then read what happened: how many affected customers contacted the business about the problem, how many used the choices offered, what they asked that the message did not answer, and how long the gap was between the business knowing and the customers being told. Compare those contacts with an earlier disruption that was not announced, in your own records; that comparison is a more honest KPIA metric tied directly to an objective. for proactive messages than any open rate. The cause of the disruption goes to the root-cause review the recovery guide describes.
A coach tour operator tells customers about a cancelled route (illustrative)
Tremblix Coach Tours announces a cancelled day before customers ask
Tremblix Coach Tours, an invented coach tour operator, learns on a Monday that a road closure will block one day's route on its 5-day scenic tour for 3 departures over the next fortnight. The operations manager, who owns disruption messages, confirms the facts that afternoon: the day's excursion cannot run, the hotel nights are unaffected, and a replacement visit to a nearby town is possible.
The booking system lists 118 passengers on the 3 departures. Cross-checking against the coach manifests finds 4 late bookings the first filter missed. That evening every passenger receives an email, and those departing within 3 days also receive a text. The subject line reads "Day 3 of your scenic tour is changing". The first paragraph says what changes and the 2 choices: take the replacement visit or receive a partial refund for that day. The reason takes one sentence, the apology one more, and the message promises an update by noon on Wednesday. It contains no offers. The tours page carries a dated notice with the same wording.
The phone team is briefed before the email is sent, with the passenger list, the 2 choices and the refund amount they can confirm. The Wednesday update goes out on time, saying the position is unchanged and the replacement visit is booked. After the last affected departure, the operator counts 27 calls about the closure from the 118 passengers. During a coach breakdown the previous summer, which affected 96 passengers and was not announced, it logged 71 calls, many of them complaints. The closing message thanks passengers and links a short form for anything still unresolved.
Common mistakes
- Waiting until everything is known. Customers notice first and the message arrives as a reply to complaints.
- Sending before anything is known. A vague message produces as many contacts as silence, and often a correction.
- Messaging everyone. Customers who were never affected are alarmed and get in touch.
- Burying what it means for the reader. Background first and the change in the third paragraph means most readers miss it.
- Offers in the notice. A discount attached to a disruption message turns a service message into a promotional one; treat any recognition as a remedy.
- No promised update, or a broken one. Customers fill silence with the worst reading.
- Support hearing about it from customers. Brief the team with the same facts before the message goes out; the service and support hub places this within routing and standards.
- No closing message. Customers left with an open problem keep checking, and keep contacting.
