The welcome flow playbook
The welcome sequence gets the highest engagement your email programme will ever see. Spending it on "here is 10% off, here is our story, here is the discount again" wastes the one moment everyone is listening.
A subscriber’s first week is when open rates are double their future baseline and buying intent is at its peak. This playbook builds the sequence that uses that window: a capture worth signing up for, messages sequenced by what the subscriber does rather than a fixed day count, and a path to first order that does not teach every future customer to wait for a coupon.
Stage 1: Capture with a reason
The flow starts before the first email. A capture that offers something specific — a genuinely useful guide, early access, or yes, sometimes an incentive — outperforms a bare "subscribe to our newsletter" box severalfold. Where the incentive is a discount, decide now how it interacts with your margin and your brand’s pricing integrity, because the welcome flow trains expectations.
A capture offer matched to the category: guides for considered purchases, incentives for impulse ones
Timing and placement tested — exit-intent and post-scroll beat instant popups
Consent and expectation set at capture: what they will get, how often
The discount decision made deliberately, with its training effect priced in
Stage 2: Sequence on behaviour, not the calendar
Day-based flows send the same third email to someone who bought yesterday and someone who never opened. Branch on the signals instead: browsed a category → speak to that category; visited pricing or shipping pages → answer the objection those pages raise; bought → exit into the post-purchase flow immediately. The skeleton is three to five messages: the promised delivery, the brand’s actual difference, proof (reviews, results), objection handling, and the close.
Stage 3: Measure to first order, revise quarterly
The flow’s metric is conversion to first order within the window — not opens. Track it per message and per branch, kill or rewrite the steps that do not move it, and re-test the incentive question with a holdout: many stores discover the discount recovers orders they would have had anyway. The welcome flow decays like every flow; a quarterly revision keeps it honest.
How it works
Fix the capture
An offer worth signing up for, placed and timed properly, with expectations set.
Build the branching sequence
Three to five messages branched on browse, objection, and purchase signals.
Measure to first order
Per-message conversion tracked, incentive tested against holdout, quarterly revision scheduled.
What you get
A capture rate worth the traffic you already pay for
Sequences that respond to what subscribers do, not what day it is
A first-order conversion number you can state, per branch
A discount policy chosen deliberately instead of inherited
Handled by these agents
Works with
Frequently asked questions
Should the welcome flow lead with a discount?
Only if the capture promised one, and even then consider placing it later in the sequence. Leading with it trains everyone — including full-price buyers — to expect it. Test with a holdout; the answer differs by store and the data settles it.
How many emails is too many?
The sequence should end when incremental conversion per message goes flat — typically three to five sends. Past that you are spending future deliverability on present hope.
What about SMS in the welcome flow?
Where consent exists, an SMS touch works best at the close — the reminder, not the essay. Channel by message job: email carries the story, SMS carries the nudge.
Which agents run this?
Retention builds and revises the flow in Klaviyo; Analyst supplies the browse-signal branching and the holdout measurement. The quarterly revision happens on schedule because it is the agent’s job, not a calendar hope.
Keep reading
Email flow automation
Complete lifecycle coverage, built once and then actually maintained.
Build an abandoned browse flow
Recover interest that never reached the cart — with intent thresholds that keep it welcome.
Customer segmentation
Segments built from purchase behaviour, not from generic engagement windows.
Optimise product titles
Titles that match how buyers search — across store, Google, and marketplace surfaces.
Put the team to work on your store
Connect your storefront, analytics, and email stack, set a goal, and let the agents run the work end to end. Start on the free plan with your own model key.