Lifecycle email & SMS

Your flows should be a third of your revenue.

Most brands built their Klaviyo flows once, at launch — often by someone who has since left the company. We rebuild the stack so the timing and the segment are predicted per customer from what they actually consume, instead of typed into a form two years ago and never touched.

Book the audit → Measure your gap →

41%

of email revenue comes from automated flows
5%

of sends is all it takes
18×

the revenue per recipient versus campaigns

Source: Klaviyo benchmark data, 183,000+ brands

The problem

Flows are the highest-leverage asset in your marketing stack and the most consistently neglected. The reason is structural: campaigns are visible — someone has to send them every week — and flows are invisible. They run whether or not they work.

So the typical $5M brand has a welcome series written before product-market fit, an abandoned cart email with one touch instead of three, a post-purchase flow that says “thanks for your order” and nothing else, and no replenishment logic at all. That brand is doing maybe 15% of revenue from flows when the category benchmark is closer to 35%.

For brands whose product gets used up and reordered, the gap is wider than in any other category — because a replenishable product has a natural rhythm that flows can be built around, and almost nobody builds around it.

That rhythm is a prediction, not a setting. A 30-day reorder email is a guess applied identically to the customer taking one capsule a day and the customer taking three, to the 60-count bottle and the 180. The data that would settle it is already in the account: the pack size on the order, the dose on the label, and the gap between that customer’s own previous orders. Fit those and you get a run-out date per customer per SKU, and the prompt lands on it instead of on day thirty.

The same is true of everything else the flows decide. A segment typed into a form in 2023 is a snapshot of who mattered in 2023; a model re-scored as the orders arrive is not. A send time chosen because someone liked Tuesday mornings is not a decision about your customer at all. None of this is content generation — nothing here writes your emails. It decides who hears from you, and when.

What we build

Nine flows on one architecture, with a single segmentation model and a fitted consumption curve underneath all of them — built, written and tested inside your account, so all of it stays yours.

Welcome & onboarding Branched on what is already knowable at signup — first product, pack size, subscription versus one-time, where they came from — as a scored split rather than a hand-kept list. Someone starting on a trial size needs a different sequence than someone reordering a refill they have bought nine times.
Abandoned checkout, cart & browse Three-touch minimum on checkout, with the gap between touches set per recipient from the hours that person has previously opened and bought in. Three-email sequences generate roughly 6.5× the revenue of single emails on the same traffic.
Post-purchase education Timed against the predicted usage curve rather than a fixed day count — day 21 for the two-a-day buyer, day 60 for the same product in a larger pack. For supplements this is the single highest-ROI flow you can build; the month-three cliff is an education problem.
Replenishment & reorder A consumption model fitted per SKU on your own data — pack size, dose or usage rate, and the real gap between each customer’s orders — rather than a flat 30-day guess. The prompt lands on the date that customer is predicted to run out, which for one SKU is a different date in different households. This is where a specialist beats a generalist agency.
Subscription upgrade paths One-time → subscription. Monthly → prepaid or annual. Offered to the customers whose reorder pattern says they will actually use the cadence, rather than to the whole list. Annual billing roughly doubles 12-month retention versus monthly.
Winback & sunset Segmented by the likely reason they lapsed — read from what they bought, when the orders stopped and what they stopped opening — rather than by days since last order, with real list hygiene so deliverability doesn’t degrade.
VIP & loyalty Recognition for the top decile by predicted value rather than trailing spend, so the customer three orders into a pattern that ends up in the top decile is found before they get there. Today they get the same emails as everyone else.
The SMS layer Postscript or Attentive, integrated with email rather than running in parallel and double-messaging people — one decision about who hears what, on which channel, at which hour.
Deliverability foundation Authentication, sending reputation, engagement-based segmentation, suppression logic. Everything above is worthless if you’re landing in Promotions — a predicted send time cannot help you in a tab nobody opens.

Process

Audit & benchmark

Every existing flow mapped, measured, and scored against category benchmarks. The order history comes out at the same time, because it is what the consumption and segmentation models are fitted on. You get the gap in dollars.

Week 1

Architecture

Flow map, trigger logic, the segmentation model and a fitted consumption curve for each core SKU — written down as what it reads (pack size, dose, order gaps, open and purchase hours), what it predicts (the run-out date, the send hour, the segment) and where it is not allowed to decide alone.

Week 1–2

Build & copy

Every flow built and written. Copy in your voice, inside your claims constraints. The model decides who receives a message and when; it never decides what the message says.

Week 2–4

QA & launch

Every branch tested with real test profiles, and the replenishment model back-tested against last year’s reorders before it is trusted with a date. Nothing goes live untested.

Week 4

Measure & iterate

Weekly reporting on flow revenue share, RPR by flow, and the gap that’s left — plus how far each predicted reorder date sat from the actual one, which is the number that says whether the model is earning its place.

Ongoing

Benchmarks — flow revenue per recipient

Revenue per recipient, by flow
Flow	Typical RPR	Top decile
Abandoned checkout	$2.96–$9.00	$28.89
Welcome series	$1.50–$7.00	—
Post-purchase	$0.47–$5.00	—
Browse abandonment	$0.40–$3.50	—
Winback	$0.07–$2.50	—
Campaign baseline	~$0.11	—

Aggregated agency and Klaviyo benchmark data. Your numbers will vary by AOV and list quality — we benchmark against your category, not against ecommerce generally.

What it costs

Flow audit, inside the Revenue Recovery Audit

Included
Full flow stack build

$6,000–$12,000
Ongoing optimization

$3,000–$8,000/mo

The audit fee is credited in full against any build you go ahead with.

Questions

Do you write the copy, or do we?

We write it. You approve it. The model decides who receives a message and at what hour — never what it says.

What is actually predicted, and what is just a rule?

Predicted: the run-out date per customer per SKU, the send hour per recipient, the segment a customer belongs in this week, and the likely reason a lapsed buyer stopped. Rules: consent, quiet hours, frequency caps, suppression and anything touching a claim you are allowed to make. Rules are cheaper, testable and impossible to talk out of their logic, so they hold every boundary that matters — the model only chooses timing and audience inside them.

How accurate is the replenishment model?

We publish no accuracy figure — metric to confirm. What we do before launch is back-test it against your last year of reorders, so you can see how far its predicted dates sit from the dates people actually reordered, per SKU. After launch that gap is reported alongside flow revenue, because it is the number that says whether the timing is working or whether the flow is just sending.

Can you work with our existing Klaviyo account?

Yes, always. We work in your account and you keep everything.

What if we use Omnisend, Customer.io or Attentive?

Fine. Klaviyo is most common in this category, but the architecture is platform-agnostic.

How long until we see revenue change?

Flow revenue moves within 30 days of launch. Replenishment timing takes a full purchase cycle to read properly.

Do you handle SMS compliance?

Yes — consent capture, quiet hours, and opt-out handling are part of the build.

Find out what you’re losing.

Before you commit to anything, we tell you exactly what you’re losing and what it costs to stop it. Two weeks. Fixed fee. Credited in full against any build you go ahead with.

Book your Revenue Recovery Audit →Or run the numbers yourself →

Fee: $1,500–$3,000, fixed
Duration: Two weeks
Credited: In full, against any build
You supply: Read access + one 45-minute call