What this playbook covers and when to run it
The playbook covers lifecycle email as a programme: deciding which journeys exist and why, building and QA-ing them, monitoring them once live, testing improvements and auditing the whole set against the customer lifecycle each quarter. It does not teach how to write a welcome series, how to build segments in a particular platform or how sender reputation works; the welcome series guide and the deliverability guide own those subjects.
Run it when email has moved beyond one-off campaigns and the business has, or wants, automated journeys tied to customer behaviour. It is also the right tool when a programme has grown organically: a dozen flows built by different people over the years, some still firing, some forgotten, and nobody sure which ones overlap. The first pass through the playbook will often mean switching things off before building anything new.
The playbook is part of the playbooks library. It connects email to the wider lifecycle marketing guide, which describes the stages journeys are mapped to, and to the customer retention strategy guide, which sets the retention goals journeys serve.
Roles and responsibilities
- Lifecycle marketer. Owns the programme: the lifecycle map, journey priorities, content and the test planThe written design of a test, fixed before launch: hypothesis, unit of randomisation, primary metric, guardrails, planned sample, duration and decision rule.. Chairs the monthly and quarterly reviews.
- Email developer. Owns the build: templates, journey logic, triggers, personalisation and rendering across clients. Runs technical QA before launch.
- Data analyst. Owns segments and measurement: segment definitions, data feeds into the email platform, holdout groupsA randomly chosen part of an audience kept from exposure to measure what exposure caused. and the analysis of journey performance and tests.
- Compliance reviewer. Owns the consent and data-use questions: who may be emailed, on what basis, with what opt-out. Reviews every new journey and any change touching consent.
The analyst role is often missing in smaller teams, and the result is segments that drift. A segment called "active customers" may have been defined years ago with a rule nobody remembers. Whoever holds the analyst role should own a short written definition of every segment the programme uses. Where a role is bought in from an agency or a freelancer instead of hired, the guide to hiring and evaluating an email marketing agency sets out what stays with the business and what to ask the provider.
Cadence, inputs and outputs
| Cadence | Owner | Inputs | Outputs |
|---|---|---|---|
| Weekly | Lifecycle marketer, email developer | Flow volumes and engagement, bounces, complaints, unsubscribes, error logs | Broken or misfiring flows fixed; deliverability issues escalated |
| Monthly | Lifecycle marketer, data analyst | Segment sizes and definitions, test results, journey conversion | Test decisions, segment fixes, the next tests scheduled |
| Quarterly | All roles | Lifecycle map, journey inventory, programme-level results | Journeys added, changed or retired; the updated inventory |
The weekly check is deliberately operational. It asks whether journeys are firing and whether mail is being delivered, not whether the copy could be better. Content and offer questions wait for the monthly review, where test results and segment health give them a proper evidence base. The quarterly audit is the only point at which journeys are added or retired, which keeps the programme from changing shape every week in response to whichever request arrived last.
The playbook in phases
Map the lifecycle
Start with the customer, not the email platform. Write down the stages a customer moves through, from first sign-up to first purchase, repeat purchase, lapse and win-backMessages aimed at customers who have stopped buying or engaging, sent once they pass their usual gap between purchases., using the stages your business actually sees. For each stage, note what the customer is trying to do, what the business wants to happen next and which signals show that someone has entered or left the stage. Then lay the existing journeys over the map. Most programmes find stages with several overlapping journeys and stages with none. This map, built with the customer journey mapping approach, is the reference every later decision is checked against, so share it with the whole team and keep it current.
Prioritise journeys
List the candidate journeys: gaps on the map, existing journeys that need rebuilding, and overlaps that need merging. Score each on reach (how many customers pass through that stage), value (what moving them on is worth, which is where estimates of LTVTotal value a customer brings over time. help), current experience (what they receive now) and effort to build. Put the scores in a table so the trade-off is visible rather than argued from memory. Pick a small number to build this quarter. Also decide what to switch off; a journey that duplicates another, or targets a segment nobody can define, costs attention and risks sending conflicting messages.
Build and QA
For each journey, the lifecycle marketer writes a short brief: trigger, entry and exit rules, suppression rules, the sequence of messages and waits, the goal and the holdout. The email developer builds it and runs technical QA: rendering across major clients, links and tracking parameters, personalisation fallbacks, trigger and exit logic with test profiles. The data analyst checks that the segments feeding the journey match their definitions. The compliance reviewer checks consent basis and opt-out handling against the direction in the ICO and FTC guidance. Nothing launches until all four have signed off, and the sign-offs are recorded against the journey in the inventory.
Launch and monitor
Launch journeys gradually where the platform allows, starting with a share of eligible customers, so a logic error affects few people. In the first weeks, check daily that the journey fires, that volumes match expectations and that nobody is receiving messages from conflicting journeys. After that, the weekly check covers every live journey: entry volumes, opens or clicks as a rough health signal, bounces, complaints and unsubscribes. A sudden change in any of these usually means a trigger broke or a data feed changed. List hygiene, including suppressing hard bounces and long-inactive addresses, follows the list hygiene procedure on its own schedule.
Test and review
Each live journey should have a test plan. Test things that could change behaviour: timing, offer, the number of messages, the content angle. Use an A/B TestControlled comparison of two variants. within the journey where volumes allow, and keep a holdout group that receives nothing so the journey's overall effect can be measured, not just the difference between versions. When the question is the number of campaign sends rather than a journey, the email frequency test gives the subscriber-level design. The analyst reads results monthly and records them. Each quarter, all roles review the whole programme against the lifecycle map: are the journeys still the right ones, do they still fit the stages, which should be rebuilt or retired? The updated inventory is the main output.
Journey inventory and ownership
The journey inventory is the programme's single source of truth. It lists every live and paused journey with the fields needed to run it: name, lifecycle stage, trigger, goal, owner, date launched, date last reviewed, current test, holdout status and sign-offs. When someone asks "what emails does a new customer get in their first month?", the inventory should answer it without anyone logging in to the platform. Broadcast sends are planned against this inventory in the email marketing calendar template, so that a promotion and a journey do not reach the same person in the same week.
| Field | Why it matters |
|---|---|
| Stage | Connects the journey to the lifecycle map and exposes overlaps |
| Trigger and exit rules | Lets anyone check the journey still fires correctly after a data change |
| Goal metric | Defines success, usually a CVRShare of visits or clicks that convert. to the next stage or a retention measure |
| Owner | Makes sure every journey has someone to notice when it breaks |
| Holdout | Shows whether the journey's effect can be measured at all |
| Last reviewed | Exposes journeys that have run untouched for too long |
Ownership matters because journeys outlive the people who built them. When a lifecycle marketer leaves, their journeys should be reassigned by name in the inventory, not inherited by whoever happens to log in next. The quarterly audit checks that every row has a current owner. The academy capstone on an ecommerce trading operations plan for a party supplies shop practises planning a lifecycle and retention programme with its journey inventory, holdouts, audit and win-back test, beside a peak and a Shopping programme in the same year.
A homeware retailer pruning before building
An online homeware retailer has 23 automated flows built over five years by three different agencies. The new lifecycle marketer maps them against a simple five-stage lifecycle and finds that new subscribers can receive messages from four flows in their first week, while lapsed customers receive nothing. The team switches off 9 flows, merges 4 into a single welcome journey with a 10% holdout, and builds one win-back journey. Three months later, unsubscribes in the first fortnight after sign-up have fallen by roughly a third, and the welcome journey shows a 6% higher first-purchase rate than its holdout.
Measuring the playbook
Measure the programme at two levels. At the journey level, look at each journey's goal metric against its holdout, which is the only honest measure of whether the journey adds anything beyond what would have happened anyway. At the programme level, look at movement between lifecycle stages over time: are more new customers making a second purchase, are fewer lapsing, is the FunnelModel of the customer journey stages. from sign-up to first order getting shorter?
Operating signals show whether the playbook itself runs: the share of live journeys with a named owner and a recent review date, the share with a holdout, the number of journeys with a current test, how quickly broken flows were noticed and fixed, and whether complaint and bounce rates are stable. A programme whose results look fine but whose inventory is out of date is one data change away from a problem nobody will notice for weeks.
For how a lifecycle programme should be judged when retention is the objective, retention through email sets the cohort KPI set and the holdout-based budget posture.
Win-back journeys need their own holdout, and the win-back offer test shows how to read a lapsed-customer offer against a message-only arm and against customers who receive no win-back contact.
Common mistakes
- Building journeys without a lifecycle map. Journeys get added to solve whatever problem is loudest, and the programme ends up with overlaps and gaps. Map first.
- No holdout. Without a group that receives nothing, a journey's reported conversions include people who would have converted anyway. Keep a holdout on every journey that can afford one.
- Segments nobody can define. A segment whose rule is unknown cannot be trusted. Write every definition down and review it monthly.
- Skipping compliance on small changes. A change to the opt-out or to the data used for targeting can matter as much as a new journey. Agree which changes need review.
- Never retiring flows. Old journeys keep sending long after their purpose has gone. The quarterly audit exists to switch them off.
- Reading opens as success. Opens are a health signal, not an outcome, and privacy features affect how they are recorded. Judge journeys on the behaviour they were built to change.
