Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

CRM & Marketing Automation › Troubleshooting

Customers Getting Too Many Marketing Messages

Why customers say a business messages them too often: no shared cap across workflows, no priority rule, texts and app messages from several teams, one switch.
Troubleshootingintermediate12 min readUpdated

First, count what one customer received from every team

Start with a real complaint, and handle the person before the pattern. If the customer asked to stop, that is an opt-out whatever words they used, and the opt-out request procedure logs it with the scope they gave and, when they asked to stop marketing altogether or the scope is unclear, suppresses them in every tool and audience, not only the channel where the request arrived. The UK ICO's direct marketing guidance, as the site's messaging guide sets it out, expects marketing to stop once a person objects.

Then rebuild that customer's last few weeks of marketing messages from the send log of every tool that can reach them: the email platform, the text and WhatsApp tool, the app messaging tool, and anything a sales or loyalty team runs on its own. Give each message one line with the date and time, the channel, the team, the workflow or campaign behind it, and whether it was promotional or a service message the customer needed. Count the promotional lines per channel and in total, and note beside each workflow the limit its own team set. A list in which every workflow kept to its limit while the total is still too high points at the rails rather than at any one team.

The customer journey mapping framework gives the list a second axis. Name the stage the customer was actually at, such as just bought, waiting for a delivery or comparing options, and place each message in the touchpoints lane under it. A map shows what happens but not how many times, so the count supplies what the map leaves out, and together they show messages written for one stage arriving at another. Read the customer's record too: whether a complaint or dispute was open, and whether they exist as more than one contact.

Some complaints belong elsewhere. The same email arriving twice is a matter for the diagnosis for subscribers receiving duplicate emails; a jump in email exits after a change of cadence, for the unsubscribe spike diagnosis; app users switching notifications off, for the page on users turning off notifications; and customers receiving sequences written for new leads, for the diagnosis for customers receiving prospect messages. This page stays with the count across channels and teams.

Diagnosis table

Diagnosis

SymptomCheckLikely causeFix
Every email workflow kept to its own limit, yet the customer's inbox total is too highWhether any condition counts a person's promotional messages in the channel before each sendNo channel-level cap checked before every sendAdd one cap per channel as a shared rail, read before every send, with service messages exempt
A flow, the newsletter and a sales sequence landed in the same few daysWhether a written priority order and mutual exclusion rules exist across flows and campaignsNo priority order or mutual exclusionWrite one priority order and enforce it with entry conditions and an active-flow flag
App messages came from marketing, product and loyalty teams in one weekWhether every tool that sends app messages reads one shared capNo single cap across every campaign and team that sends app messagesSet one cap per person per period that every sending tool reads
Texts or WhatsApp messages arrived from several flows in a weekWhether the cap sits in each messaging workflow or across the channelMessaging capped per workflowMove the cap to the channel, before every promotional send
Customers ask to hear less, then turn everything offWhether the app and email offer a choice by kind of messageOne switch for every kind of messageOffer preferences by topic, read by every sending tool
The loudest complaints come from customers with a problem already openWhether sends read any field that marks an open complaint or disputeNo hold while a complaint is openAdd a temporary hold to the suppression rail, released by rule

Fix, verify and prevent each cause

Each workflow restrained, but no cap across the channel

The automation governance guide puts it plainly: message frequency is where workflows that are each reasonable add up to an unreasonable experience, and a cap on marketing messages per person per period, checked at the channel level before every send, is the one control with the whole picture in view. A limit inside each workflow cannot see the others.

Fix: add the cap as a shared rail beside the suppression list, read before every promotional send in the channel, with transactional and service messages exempt, and treat any change to it as a reviewed change to a shared asset. A cap that counts per contact can also miss a person held as two records, and the duplicate contacts diagnosis finds the entry point that created the second one.

Verify: put a seed contact into several flows and a campaign at once, and confirm that the channel delivers no more than the cap and that service messages still arrive.

Prevent: give the cap a name that marks it as shared and protected, and make reading it a condition of switching on any new sending workflow.

No priority order or mutual exclusion between flows and campaigns

A cap decides how many messages go; a priority order decides which. Without one, whichever workflow fires first takes the slot, and a newsletter can crowd out the replenishment reminder the customer wanted. The governance guide decides, for one contact, which workflows may not run side by side and which one wins: a demo request ahead of a nurture, a cart flow ahead of a newsletter, a service message ahead of everything. The lifecycle marketing guide points the same way, treating scheduled sends such as the newsletter as something laid over the triggered flows, not a substitute for them.

Fix: write one priority order covering flows, campaigns and sales sequences, record each workflow's place in it in the inventory, and where the platform has no built-in priority, enforce it with entry conditions and a flag set while a higher-priority flow is running.

Verify: a seed contact that qualifies for a flow, the newsletter and a sales sequence on the same day receives the one the order puts first, and the others wait or drop as written.

Prevent: place every new flow or campaign in the order before it goes live, and include sales and account teams' sequences in the same list.

App messages sent by several teams with no single cap

In an app, product nudges, marketing offers, loyalty rewards and service alerts can come from different teams through different tools, each modest on its own. The push and in-app messaging guide asks for one limit for each person over a set period, covering every campaign and every team able to send, with service messages exempt and promotions at the back of the queue. A push notification interrupts the person wherever they are, as the push guide describes it, so each one has to earn its place in the total.

Fix: add every tool that sends app messages to the customer history the first section rebuilt, so app messages are counted beside email and texts, and hold one cap, agreed between the sending teams, that every sending tool reads before each send; where the evidence is app users switching notifications off rather than complaints, the notifications diagnosis linked above covers the same rail from the app's side.

Verify: rebuild the complaining customer's history for a comparable period after the change, and confirm that app messages stay at or under the cap.

Prevent: record the app cap in the same inventory as the email and text caps, so a team changing one sees the others.

Texts and WhatsApp messages capped per workflow

Texts and WhatsApp messages reach the phone at once, which is why the messaging guide holds them to stricter rules than email. The guide to SMS and WhatsApp automation describes someone enrolled in a cart reminder, a renewal flow and a restock alert at the same time and getting several messages in one week, each workflow restrained on its own terms, and names one cap for the whole channel, checked before each promotional send, as the dependable control. It also notes that the CTIA messaging principles expect the opt-in to state the expected message frequency.

Fix: move the limit out of each messaging workflow into one channel cap, set no higher than the frequency the opt-in wording promised, with transactional messages exempt and correctly classified.

Verify: compare the opt-in wording with the cap, and run a seed contact through every messaging workflow at once.

Prevent: keep the opt-in wording and the cap on the same record, so neither changes without the other, and drop people who have not responded to several promotional texts to service messages only, as the guide advises.

One switch for every kind of message

When the only choice is everything or nothing, a customer tired of offers loses the order updates too, and the business loses the channel that carried them. The tutorial on topic preferences for notifications sets out the alternative: topics named in the customer's language, transactional and promotional kept apart, promotional topics counted towards the cap, an owner for each topic, and every sending tool reading the same preferences before each send. The tutorial covers app topics and notes that email and text preferences sit beside them, each with its own consent record. For app users who switch notifications off altogether, the notifications diagnosis linked above covers the same missing topics from the app's side.

Fix: inventory every message type, group them into topics, offer the same choice in the email preference centre, and make every tool check the preferences before each send.

Verify: turn off one promotional topic for a seed contact in the app and in the email preference centre, then rebuild a week of that contact's messages as the first section did; nothing on that topic arrives in either channel, and order updates still do.

Prevent: when a topic is added in the app, add the matching choice to the email preference centre in the same change.

Promotions still reaching customers with an open complaint

A customer waiting on a late delivery or disputing a charge passes every consent and suppression check and still receives the weekly offer, at the moment the relationship is already strained. The tutorial on holding marketing while a customer's issue is open adds a hold: a temporary flag on the record, giving the reason and when it began, set by whoever learns of the issue, read by every send and audience, and lifted by a rule that someone owns, while the service messages the customer needs keep running.

Fix: add the hold to the shared suppression rail, have the complaints log set it, and carry it to text, app and sales sequences and to uploaded customer lists.

Verify: put a seed contact on hold and send it through every workflow on the list: no promotion should reach it, and every service message should.

Prevent: keep the hold separate from an objection to marketing, because the hold ends when the issue closes and an objection does not.

Prevention: one cap per channel, one priority order and preferences by topic

Every cause above comes from judging messages one workflow or one team at a time. Prevention moves the judgement to the level the customer experiences: the channel and the person. Hold one promotional cap for each channel as a shared rail, read before every send and owned by one person, with service messages outside it. Hold one priority order across flows, campaigns and the sequences sales and loyalty teams run, written into the inventory. The workflows hub lists a contact receiving several sequences at once among its common problems, and both rails exist to stop it. The customer contact policy template writes every rail on this page into one document that each sending team reads, from the caps and the order to topics and holds.

Give customers a way to hear less. Topic preferences in the app and the email preference centre, read by every sending tool, turn a complaint into a choice. Treat timing as a rail too: quiet hours held in each recipient's local time as a system rule for texts and push, with service messages let through, as the push and messaging guides set them, while the legal review for each market sets any hard limits. For texts, the frequency stated at opt-in is part of the same rule.

Keep the hold for customers with an open issue joined to the complaints log, so it starts and ends without anyone remembering to set it. When a request to stop arrives, by any route, it goes through the opt-out procedure rather than a reply from whoever saw it.

Then watch the total, not the workflows. The governance guide's volume alarms flag a workflow that sends far more than usual; add a regular look at the customers who received the largest number of promotional messages across all channels, and rebuild one of their histories the way this page began. Make reading the caps, the priority order, the preferences and the hold a condition for any new sending tool or team, and include them in the quarterly audit. When customer service or the social team hears the complaint again, the first question is which of these rails the customer slipped past. The automation troubleshooting section covers the neighbouring problems in workflows and the CRM.

A pet supplies retailer counts one customer's fortnight (illustrative)

Illustrative example — not a real result

Ivoldraxi Pet Supplies

Ivoldraxi Pet Supplies, an invented pet supplies retailer, logged 23 contacts in a month from customers saying it messaged them too often. The CRM lead rebuilt one complaining customer's fortnight from 3 send logs: 2 newsletters, 3 replenishment reminders and 4 browse-flow emails from the email platform, 4 texts from a separate promotions tool, and 5 app messages from the loyalty team, 18 promotional messages in 14 days. Each team had kept to its own limit: no email flow sent more than 4, the promotions tool kept to 2 texts a week and the loyalty team to 3 app messages a week. The same customer had an open complaint about a late delivery for 11 of those 14 days, and 2 of the texts arrived after 9 pm.

The retailer set a promotional cap per channel checked before every send: 3 emails, 1 text and 2 app messages a week, with order and delivery messages exempt. It wrote one priority order with replenishment above browse and the newsletter last, gave the app and the email preference centre 4 topics, added quiet hours from 8 pm to 9 am local time for texts, and had the complaints log hold promotions while a complaint was open. Replayed against the new rules, the same fortnight would have carried 4 promotional messages, none while the complaint was open. In the following month, customer service logged 7 contacts about message volume.

Common mistakes

  • Judging each workflow on its own. Every flow passes its own test while the customer receives all of them.
  • Counting email only. Texts, WhatsApp and app messages from other teams land on the same person.
  • Capping service messages. Order and delivery updates are what customers expect; keep them outside the cap.
  • Leaving sales and loyalty sequences out of the priority order. They reach the same customers as marketing.
  • Offering one switch. People who want fewer offers lose order updates, or stay and keep complaining.
  • Treating a complaint about volume as a request for nothing. When it is unclear whether they want to stop or to hear less, ask once, hold all marketing to them until they answer, and act on the answer in every channel it covers.

Frequently asked questions

How many marketing messages is too many?

The site's guides set no number, because the right limit depends on the channel, the business and what customers agreed to when they signed up. Write down a cap for promotional messages in each channel, check it before every send, and keep service messages outside it; for texts, keep it no higher than the frequency the opt-in wording described. Then let customers show you whether it is right: complaints, opt-outs and notifications switched off after busy weeks are the signs that the cap sits too high.

Do order updates and delivery messages count towards the cap?

No. Messages a customer needs because of something they did, such as order confirmations, delivery updates and account notices, sit outside a promotional cap, and promotions give way to them when both qualify. The exception is a service message that carries a sales pitch: a delivery update with an offer panel counts as marketing for this purpose, so either take the offer out or count the message.

A customer said on social media that we message them too much. Is that an unsubscribe?

If they asked you to stop, yes. A request to stop marketing counts whatever words are used and wherever it arrives. It is logged with the scope the customer gave; a request to stop all marketing is applied in every channel and tool, not only the one where it arrived, and when the scope is unclear the business asks once and stops everything in the meantime. If they asked for fewer messages rather than none, reply with the ways to choose less, such as topic settings in the app or an email preference centre, and use their message history to find which rail was missing so the next customer is not pushed to the same point.

Terms used on this page

Frameworks

Sources

  1. Direct marketing guidance — Information Commissioner's Office (UK)
  2. CTIA messaging principles and best practices — CTIA

Related