Skip to content

Conversion is Powered by Oxify

Conversion
EmailSMSSubscriptionsDeliverability

Retention & Lifecycle

The ecommerce retention agency measuring what email and SMS actually add

Lifecycle flows segmented by real behaviour, deliverability managed rather than assumed, and holdout groups that separate incremental revenue from the revenue your platform takes credit for.

  • Holdout testing within segments — incremental, not attributed
  • Deliverability fixed before content, because spam placement wastes both
  • SMS with its own strategy, not a duplicate of the email calendar
  • Subscription churn treated as an experience problem, not a discount problem
Conversion service illustration in brand blue and green
Explore:Ecommerce MarketingShopify MarketingShopify CROGrowth MarketingShopify PPC

Why it matters

Retention is where ecommerce economics are decided

Retention is where ecommerce economics are actually decided. Acquisition costs have risen steadily across paid channels, and a brand whose customers buy once is left buying every sale at full price. A brand whose customers buy three times has a fundamentally different business, and usually a fundamentally different tolerance for acquisition cost.

Most retention programmes we inherit are not missing flows. They have plenty of flows. What they are missing is segmentation that reflects what customers have actually done, testing beyond subject lines, deliverability hygiene, and any honest measurement of whether the revenue being credited to email would have arrived anyway.

That last point matters more than it is usually given credit for. Retention platforms attribute generously by design, and abandoned cart flows in particular take credit for a great deal of purchase intent that already existed. Holdout testing is the only way to see the difference, and the difference is often large enough to change where you would spend the next pound.

We would rather run six well-segmented, properly tested flows with clean deliverability than twenty that were built once and never revisited — and we would rather report a smaller incremental number than a larger attributed one.

What we usually find

The patterns behind a retention programme that has plateaued

Almost never a shortage of flows. Nearly always segmentation, deliverability, and measurement.

Flows that treat every customer identically

The same welcome series to a bargain hunter and a high-value repeat buyer, the same winback to someone who bought once and someone who bought eight times. Segmentation by behaviour is the difference between a flow and a broadcast.

Deliverability quietly degrading

Sending to unengaged contacts to keep reach numbers up, until inbox placement falls and your engaged customers stop seeing you too. This is usually discovered months after it starts.

Open rates still being used as a metric

Apple's Mail Privacy Protection inflates opens for a large share of consumer email. Reporting on them is misleading; triggering automation off them is actively harmful.

SMS used as a second email channel

The campaign calendar duplicated to SMS, at much higher cost and much lower tolerance for irrelevance. SMS unsubscribes are effectively permanent.

Attributed revenue reported as incremental

Retention platforms credit purchases that would have happened regardless, particularly in high-intent flows. Without holdouts nobody can say what the programme actually added.

Subscription churn in the first three cycles

Usually caused by cadence mismatch, unexpected charges, or an inflexible management experience rather than the product. Discount save offers mask it and train deliberate churn.

Capabilities

What we run

Email, SMS, loyalty, and subscriptions as one lifecycle programme rather than separate channels with separate calendars.

Lifecycle flow architecture

Welcome, abandoned checkout, abandoned cart, browse abandonment, post-purchase, winback, replenishment, back-in-stock, and review requests — segmented by behaviour and purchase history.

Segmentation strategy

Segments built on what customers have actually done — value, frequency, category affinity, engagement recency — rather than on when they signed up.

Campaign calendar

A planned calendar aligned to trading, launches, and seasonality, with sending volume managed against engagement rather than maximised.

SMS programme

Time-sensitive, high-value messaging with its own strategy rather than a duplicate of the email calendar, and compliance handled properly.

Deliverability management

Authentication with SPF, DKIM and DMARC, sunset policies, engagement-based sending, and inbox placement monitoring rather than assuming delivery.

Holdout and incrementality testing

Randomised holdouts within segments to measure what the programme actually adds, separate from what the platform attributes to it.

Subscription retention

First-three-cycle churn analysis, cadence flexibility, pause and skip options, and pre-billing notification — before reaching for discount save offers.

Loyalty programme design

Structures that reward the behaviour you actually want, rather than discounting purchases that would have happened anyway.

List growth

Capture that does not damage the on-site experience, with the offer and timing tested rather than assumed.

Content and creative

Email and SMS content written in your brand voice, working from your guidelines and assets rather than generic templates.

Platform migration

Moves between retention platforms with flow logic, segments, historical data, and deliverability reputation carried across deliberately.

Reporting on repeat rate

Repeat purchase rate, time between orders, and revenue per recipient, alongside the platform's attributed figures and an explanation of the gap.

Do you know how much of your email revenue is incremental?

Most brands do not, because the platform reports attributed revenue. A holdout answers it, and the gap is usually larger than expected.

30-minute call · No commitment · Custom quote in 48 hours

Process

How we take over a retention programme

Conversion's four-stage growth process visualized in brand blue and green
01

Programme audit

Existing flows, segments, sending patterns, deliverability health, and platform configuration. Plus the commercial baseline: repeat purchase rate, time between orders, and customer value distribution.

02

Deliverability remediation

Authentication, sunset policy, and engagement-based segmentation. Done first, because improving content that lands in spam achieves nothing.

03

Segmentation rebuild

Behavioural segments reflecting value, frequency, category affinity, and engagement — the foundation everything else depends on.

04

Core flow rebuild

The six flows that carry most of the revenue, properly segmented, with content written in your voice rather than adapted from a template.

05

Measurement setup

Holdout groups established within segments so incremental revenue can be distinguished from attributed revenue from the start.

06

SMS and channel expansion

SMS introduced with its own strategy where the economics support it, alongside loyalty or subscription work where relevant.

07

Testing programme

Structured testing on offers, timing, segmentation, and content — measured on conversion and revenue per recipient rather than opens.

08

Ongoing optimisation

Campaign calendar execution, flow iteration, deliverability monitoring, and periodic holdout re-measurement as the programme changes.

How we differ

Two ways to run retention

The right-hand column is the standard model. It produces good-looking reports and makes it hard to know what is actually working.

FeatureHow we workTypical retention management
Primary metricRepeat purchase rate and incremental revenueOpen rate and attributed revenue
MeasurementRandomised holdouts within segmentsPlatform attribution
Flow countFewer, properly segmented and testedMany, built once
DeliverabilityActively managed with sunset policiesAssumed until it breaks
SMS strategySeparate strategy for time-sensitive valueEmail calendar duplicated
Subscription churnCadence and experience firstDiscount save offers
ContentWritten in your brand voiceTemplated across clients

Sectors

Where retention economics differ

Purchase frequency and replenishment behaviour vary enormously, and the programme should reflect that rather than a category benchmark.

Beauty and personal care

Predictable replenishment cycles and strong routine-building potential. Among the categories where retention economics are most favourable.

Health and supplements

Subscription-led, where first-three-cycle churn largely determines whether the unit economics work at all.

Food and beverage

High purchase frequency and strong habit formation, with replenishment timing as the dominant lever.

Fashion and apparel

Seasonal buying with category affinity signals that make segmentation unusually productive, complicated by returns behaviour.

Pet products

Consumable, frequent, and emotionally engaged — with lifecycle stage as a genuinely useful segmentation dimension.

Home and furniture

Long purchase cycles where retention is about accessories, referrals, and staying present rather than driving repeat purchase directly.

Coffee and specialty consumables

Among the strongest subscription categories, where cadence flexibility matters more than discounting.

Electronics and accessories

Accessory attach and upgrade cycles, with post-purchase education reducing both returns and support volume.

Multi-brand groups

Cross-brand customer data and the genuinely difficult question of when cross-promotion strengthens the relationship and when it dilutes it.

Deliverables

What you get

Scope varies by list size and channel mix. This is the shape of a full retention programme.

Everything included in a Conversion engagement, laid out as tangible deliverables

Programme

  • Lifecycle flow architecture with behavioural segmentation
  • Campaign calendar aligned to trading
  • SMS programme with its own strategy
  • Loyalty or subscription programme design where relevant
  • List growth capture tested rather than assumed

Deliverability

  • SPF, DKIM, and DMARC authentication
  • Sunset policy for unengaged contacts
  • Engagement-based sending segmentation
  • Inbox placement monitoring
  • Bulk sender requirement compliance

Measurement

  • Holdout groups within segments
  • Incremental versus attributed revenue reporting
  • Repeat purchase rate and time between orders
  • Revenue per recipient by flow and segment
  • Subscription churn by cycle

Content and testing

  • Email and SMS content in your brand voice
  • Structured test plan on offers, timing, and segmentation
  • Test results with decisions recorded
  • Platform migration support where needed

Why Conversion

How we work

We report incremental, not attributed

Holdouts within segments, so you can see what the programme actually added. This usually makes our numbers look smaller than the platform's and makes them worth acting on.

Deliverability before content

Improving emails that land in spam achieves nothing. Authentication, sunset policies, and engagement-based sending come first even though they reduce reach numbers initially.

Fewer flows, better segmented

Six well-segmented and tested flows reliably beat twenty built once. We will usually recommend consolidating before adding.

SMS with its own strategy

Higher cost and far lower tolerance for irrelevance mean SMS needs its own plan. Duplicating the email calendar burns a channel you cannot easily rebuild.

Subscription churn treated as an experience problem

Cadence, flexibility, and billing clarity before discount save offers, which mask the cause and train customers to churn deliberately.

Content that sounds like you

Retention content that reads as generic erodes the relationship it exists to build. We work from your voice and guidelines, not a template library.

Questions

Retention marketing, answered

What does an ecommerce retention agency do?
A retention agency owns the revenue that comes from customers you already have — email, SMS, loyalty, subscriptions, and the lifecycle programme that connects them. In practice that means automated flows triggered by customer behaviour, a campaign calendar, segmentation, and the deliverability work that determines whether any of it reaches an inbox. The distinguishing feature of good retention work is that it is measured on incremental revenue and repeat purchase rate rather than on open rates, which are close to meaningless as a decision metric since Apple's Mail Privacy Protection began pre-loading images.
How much of our revenue should come from email and SMS?
Commonly cited figures put it between 20% and 30% for established DTC brands, but the number is less informative than it looks and depends heavily on category, purchase frequency, and how long you have been building the list. A consumable product with a 30-day replenishment cycle should expect a far higher share than a furniture brand where customers buy once every several years. A more useful question than the percentage is whether your repeat purchase rate is improving, and whether the retention revenue is genuinely incremental rather than sales that would have happened anyway.
Which flows matter most?
For almost every store: welcome, abandoned checkout, abandoned cart, browse abandonment, post-purchase, and winback. Abandoned checkout is usually the highest revenue per recipient because intent is highest, and welcome is usually the highest total because it reaches everyone. Beyond those six, value depends on your model — replenishment flows for consumables, review request flows where social proof drives conversion, back-in-stock for constrained inventory. Brands frequently build twenty flows and would be better served by six that are properly segmented and actually tested.
Is SMS worth it?
Where the economics work, yes, and it is easy to get wrong. SMS has much higher open and click rates than email and much higher cost per message, plus a far lower tolerance for irrelevance — people unsubscribe from SMS quickly and do not come back. It works best for time-sensitive, high-value messages: abandoned checkout, restocks, launches, and shipping updates. It works badly as a second email channel sending the same campaign calendar. Compliance also matters more than in email, with real legal exposure under TCPA in the US and equivalent regimes elsewhere.
What about deliverability?
It is the constraint most brands discover only after it breaks. Sending to unengaged subscribers to inflate reach damages sender reputation, and once inbox placement degrades your best customers stop seeing you alongside everyone else. The practical work is sunset policies for unengaged contacts, proper authentication with SPF, DKIM and DMARC, segmentation by engagement rather than blasting the full list, and monitoring placement rather than assuming delivery equals inbox. Google and Yahoo's bulk sender requirements have made this materially less optional than it was.
Why are open rates no longer reliable?
Apple's Mail Privacy Protection pre-loads tracking pixels for Apple Mail users regardless of whether the message was opened, which inflates open rates and makes open-based segmentation and automation unreliable. Since Apple Mail represents a large share of consumer email, this affects most ecommerce lists substantially. Click-through, conversion, and revenue per recipient remain trustworthy. Any agency still reporting open rate as a headline metric, or triggering automation off opens, is working with a broken instrument.
How do you measure whether retention marketing is actually working?
By repeat purchase rate, time between orders, and revenue per recipient — with holdout groups where list size permits. Attributed revenue in a retention platform is systematically generous: it credits the email for purchases from customers who were already going to buy, particularly in flows like abandoned cart where intent was already high. Holdout testing, where a random slice of a segment receives nothing, is the only reliable way to separate incremental revenue from recorded revenue. We run them because the difference is frequently large.
Klaviyo, or something else?
Klaviyo is the default for Shopify for good reasons: deep native integration, strong segmentation, and combined email and SMS. It is not the cheapest, and at large list sizes the cost becomes material enough to warrant comparison. Alternatives worth considering depending on circumstances include Omnisend, Attentive for SMS-led programmes, and Braze for genuinely enterprise multi-channel requirements. That said, platform choice is rarely the constraint on results. We have seen far more programmes limited by segmentation and content than by tooling.
Can you help with subscriptions?
Yes. Subscription work splits into acquisition — how the offer is presented on the PDP and in checkout — and retention, which is where most of the value actually sits. Churn in the first three cycles is the number that determines subscription economics, and it is usually driven by cadence mismatch, unexpected charges, or an inflexible management experience rather than by the product. Pause and skip options, clear pre-billing notification, and flexible cadence typically outperform discount-based save offers, which train customers to churn deliberately.
How long before retention work shows results?
Flow improvements can show up within weeks, because they are triggered continuously and you are improving something already running. List growth and segmentation compound over months. Repeat purchase rate, which is the metric that actually matters, moves on the timescale of your purchase cycle — if customers buy every four months, you need at least two cycles before the trend means anything. Brands with long purchase cycles should expect to wait, and should be sceptical of anyone promising otherwise.
We already have flows set up. What would you change?
Usually not the number of them. The recurring problems we find are flows sending the same content to everyone regardless of purchase history or engagement, no testing beyond subject lines, unengaged subscribers still receiving mail and damaging deliverability, SMS duplicating the email calendar, and reporting on attributed revenue with no holdout. A programme with six well-segmented, tested flows and clean deliverability reliably outperforms one with twenty that were built once and never revisited.
How do you work with our brand and content team?
Retention content has to sound like the brand or it erodes the relationship it is meant to build. We work from your brand guidelines and existing assets, and where there is an internal content team we usually own the strategy, segmentation, and testing while they own voice and creative. The arrangement that fails is an agency producing generic templated campaigns that read like they were written for any brand, because subscribers notice quickly and the unsubscribe rate tells you.

Related

Explore related services

Retention works best alongside the acquisition and conversion work feeding it.

Ecommerce Marketing Agency

Acquisition and retention run as one programme.

Explore

Shopify Marketing Agency

Full-funnel growth across paid, organic, and lifecycle.

Explore

Shopify CRO Agency

Converting the traffic before retention has to work harder.

Explore

Shopify PPC Agency

Acquisition economics that improve as repeat rate rises.

Explore

Amazon CRO Agency

Marketplace conversion, and moving those buyers into an owned relationship.

Explore

Growth Marketing Agency

Experimentation across the whole customer lifecycle.

Explore

Next step

Tell us what your repeat purchase rate looks like

A 30-minute call, a look at your current flows and deliverability, and an honest read on where the incremental revenue actually is.

No credit card · No commitment · Pay only on results