Remarketing
Remarketing is the practice of re-engaging people who already interacted with your business, often through email and ad sequences, to bring them back toward a conversion.
Key takeaways
- Remarketing works from owned contact records, so every message can be personal.
- Every trigger needs an exit condition, or contacts receive irrelevant follow-ups.
- Response decays sharply as time since the last interaction increases.
- Over-sending damages deliverability, which harms every later campaign.
- Suppression and consent checks must run before each send, not after.
In depth
Remarketing runs off records you own. Each contact carries a history: what they downloaded, which quiz they finished, whether they replied. That history is what the system queries when it decides who to contact next. A trigger fires on an event or a date, a filter checks the contact still qualifies, and a message goes out. Because the identity is known, the same person can be reached by email, phone or an uploaded ad audience under one rule.
Response falls as the gap since the last interaction grows, and it falls faster than most senders expect. It also falls when the message ignores what the contact already did. Sending more often raises short-term volume and lowers long-term engagement, which the inbox notices: repeated ignored mail pushes future messages into spam folders. The trade-off is therefore not reach against frequency but this campaign against the deliverability of every campaign after it. Good programmes protect that long-term asset first.
A working programme is a small number of triggers, each with an exit condition. Someone who booked a call leaves the nurture sequence immediately. Someone inactive for six months enters a win-back with a different tone and a single ask. Where a scorecard already scored the contact, the tier decides the sequence: a high scorer gets a short path to a conversation, a low scorer gets education until something changes. Suppression rules run before every send.
Remarketing is bounded by the list you already have, so it cannot grow a business on its own; a shrinking list produces shrinking results no matter how good the sequences are. Consent rules limit who may be contacted and for how long, and they differ by country and channel. Some contacts are simply gone: a changed job, a solved problem, a competitor already chosen. Continuing to message them costs sender reputation and buys nothing at all.
Example in practice
How to measure it
Judge sequences on the action they were built for, not on opens. For a booking sequence the number is meetings booked divided by contacts enrolled; for a win-back it is contacts who became active again. Track these per trigger rather than in aggregate, because one strong sequence usually hides several that do nothing and should be switched off.
Watch list health alongside campaign results. The share of the list active in the last ninety days, the unsubscribe rate per send, and the hard bounce rate together show whether the asset is growing or eroding. A campaign that hits its target while pushing the active share down has borrowed from next quarter. Compare re-engaged contacts against the cost of acquiring new ones to see what the programme is really worth.
Common mistakes
The most common failure is a sequence with no way out. A contact books a demo on Tuesday and still receives Thursday's message asking them to book a demo, which tells them nobody is paying attention. Every sequence needs an exit condition tied to the outcome it is chasing, checked at send time rather than at enrolment, and a global suppression list for anyone in an active sales conversation.
The second is mailing the whole list to hit a number. A quarterly push to every address, including people who have ignored a year of messages, produces a spike in unsubscribes and a lasting drop in inbox placement. Segment by recent engagement, mail the active portion regularly, and give the dormant portion one deliberate win-back before removing them. A smaller list that reads its mail is worth more than a large one that does not.
Frequently asked questions
Is remarketing the same as retargeting?
They are closely related and often used interchangeably, but remarketing usually implies channels where you have a known contact, like email, while retargeting refers to anonymous pixel-based ads. Most teams run both together for full coverage.
What triggers the best remarketing sequences?
Specific funnel actions, such as abandoning a quiz or viewing a result without booking, make the strongest triggers. Behavior-based messages outperform generic 'we miss you' broadcasts because they reference what the person actually did.
How does remarketing help a quiz funnel?
A quiz captures rich intent data, and remarketing reuses it to send tailored follow-ups based on each lead's score and answers. This turns one-time quiz takers into nurtured, sales-ready contacts over time.
How many messages should a remarketing sequence contain?
Enough to make the case once, which is usually three to five for a single ask. Each message should carry a new argument or a new piece of evidence; if you cannot write a fourth without repeating the third, the sequence is finished. Spacing matters more than count: daily is intrusive for a considered purchase, monthly is forgettable for an urgent one.
When should I stop contacting a dormant contact?
After one honest win-back attempt that acknowledges the silence and offers an easy way to leave. If nothing comes back, move them to a suppressed list rather than deleting them, so the record stays for reporting and for a future re-permission campaign. Keeping unresponsive addresses in regular sends costs deliverability for everyone else on the list.
What actually works in a win-back campaign?
Naming the gap directly tends to beat a discount. A message that says how long it has been, restates what changed since, and asks one clear question performs better than a generic offer. Give a visible way to reduce frequency instead of leaving entirely, because many dormant contacts want less mail rather than none. Send it from a person, not a brand address.
What consent do I need to remarket to a contact?
It depends on jurisdiction and channel, so treat this as a legal question rather than a marketing one. In general you need a lawful basis for the contact, a record of when and how it was obtained, a clear unsubscribe path in every message, and honouring of that request quickly. Storing the source of each contact makes an audit answerable rather than alarming.
Should remarketing run over email or ads?
Use both for different jobs. Email carries detail, links and a direct reply, and it costs almost nothing per contact. Ads reach people who never open mail and can keep a brand present while a decision is pending, but they cannot say much. Run them from the same segments so a contact does not receive contradictory messages in the two channels.
How personal can a remarketing message be before it feels intrusive?
Reference what the person told you, not what you inferred about them. Repeating an answer they gave in a form or a quiz reads as attentive; mentioning pages they browsed or how long they lingered reads as surveillance. The safe test is whether you could say the sentence out loud on a phone call without an apology attached.