The Retention Hierarchy: Why Your Retention Marketing Isn't Working

The retention hierarchy is the order in which five factors decide whether a customer comes back: product, portfolio, acquisition quality, customer experience and, last, the retention system. When retention marketing is not working, it is almost always because one of the first four is broken and the fifth is being asked to compensate.

Short answer: before you add another Klaviyo flow, rebuild the loyalty program or brief an agency on win-back campaigns, work down the hierarchy and find the constraint. Retention marketing amplifies the value that already exists in the brand. It does not create value that is missing.

The clip above starts at the retention hierarchy chapter of our ecommerce retention masterclass. This post is the written version, with the diagnostic questions we use for each level.

Why the hierarchy exists

We work with Shopify brands every week whose first instinct when repeat purchase rate drops is to send more email. The assumption underneath is that a retention problem is a messaging problem. Usually it is not. Whether a customer buys again is decided by whether the product was good, whether there is anything else to buy, who the customer was in the first place and how they were treated. The email arrives after all of that has happened, and the maths of retention versus acquisition only works in your favour if the retention system has something real to amplify.

So we rank the factors. A failure higher up cannot be fixed lower down.

The five factors, in order

1. Product

If the product is not good, the customer will not come back, and no win-back campaign will change that.

What it looks like when this is the problem: repeat purchase rate is low across every channel and every entry product; reviews mention quality, fit or effectiveness rather than delivery; customers who do return buy a different product rather than the same one again.

Diagnostic questions:

• What do the one and two star reviews actually say, and are they about the product or the service?
• Do customers who bought your hero product reorder it, or do they reorder anything at all?
• Has repeat rate changed since a formulation, supplier or manufacturing change?

What fixes it: the product. Reformulate, change supplier, improve fit, fix the defect. Pause retention spend on the affected line until then; every email about a disappointing product is a reminder of the disappointment.

2. Portfolio

Can the customer buy anything else? If there is nothing else to sell them, there is no reason to return, however good the first purchase was.

What it looks like when this is the problem: a strong review score with a weak repeat rate; a catalogue of one or two products, or a range where every item solves the same problem; customers who come back once and then run out of things to try.

Diagnostic questions:

• After the first order, what is the natural second product, and does it exist?
• What share of second orders are the same product again versus something new?
• Are there collections, bundles or complementary lines you can cross-sell, or is every campaign the same handful of SKUs?

What fixes it: range. Complementary products, refills and consumables around a durable hero product, bundles that introduce a second line, collections that give a customer a path through the catalogue. Until that exists, the time to second purchase will stay long no matter how well the email is written.

3. Acquisition quality

Not every customer is the same. A customer who bought at full price for $150 has a vastly different lifetime value from one who only bought a $40 item in the Black Friday sale. Both count as one new customer on the acquisition dashboard. Only one of them is likely to come back at full price.

What it looks like when this is the problem: repeat rate looks fine in some months and terrible in others; customers acquired on a discount code return only when there is another one; retention flows perform well for organic customers and badly for paid ones.

Diagnostic questions:

• Split repeat purchase rate by acquisition channel and by first-order discount. How different are the segments?
• What percentage of new customers were acquired at full price?
• Which channel produces the customers who reach your top VIP tier?

What fixes it: acquiring the right customers, not more of them. Shift budget towards the channels and offers that produce full-price buyers, and feed tier and repeat data back to the acquisition team so they see the lifetime value of each source, not just first-order return on ad spend.

4. Customer experience

Delivery, service, support, replacing a faulty item. A bad customer experience produces bad reviews and a negative reputation, and it hits the bottom line even when product, portfolio and acquisition are all excellent.

What it looks like when this is the problem: reviews praise the product but complain about shipping, packaging or support; a spike in support tickets precedes a drop in repeat rate.

Diagnostic questions:

• What is the repeat rate of customers who raised a support ticket compared with those who did not?
• How long does a delivery take, and how long does a replacement take?
• Does your support team know who the high-value customers are when they open a conversation?

What fixes it: operations. Faster fulfilment, honest delivery estimates, a replacement policy that resolves problems in one contact, and surfacing customer value to the support team so a VIP is not treated like an anonymous first-timer. At this level a good customer retention strategy is mostly an operations document.

5. Retention system

Email, SMS, the loyalty program. Most people think this is the only factor that matters. It is the least important of the five.

That is not the same as saying it does not matter. A well-built retention system is the difference between a brand that captures the value in the first four levels and one that lets it leak away. But it can only work with what it is given. It cannot make a customer reorder a product they did not like, from a catalogue with nothing else in it, after a delivery that arrived late.

What it looks like when this is genuinely the problem: reviews are strong, the range has an obvious second purchase, customers are acquired at full price and support is good, yet second orders still arrive slowly or not at all. Customers are not reminded, not given a reason, and not recognised when they return.

Diagnostic questions:

• Does a customer receive anything between their first order and the point at which you would expect the second?
• Is there a specific reason to return (a threshold to reach, a product they have not tried, early access they will lose), or just points?
• Does a tenth-order VIP get the same experience as a first-order customer?

What fixes it: a retention system designed around a behaviour, which is the last section of this post.

The $150 customer and the $40 customer

The acquisition quality level deserves one more example, because it is the one brands most often miss.

Two customers arrive in the same week. The first paid $150 for a full-price order after reading reviews and browsing for a week. The second paid $40 for a single item in the Black Friday sale and has not visited since. On a new-customer report they are identical. In reality the first customer has shown willingness to pay, interest in the range and a habit of buying at full price. The second has shown that they respond to a sale.

If your retention marketing treats them the same, it will fail for at least one of them. Send both a full-price launch email and the second customer ignores it. Send both a discount and you have taught the first customer to wait for the next one. Knowing who you are acquiring is the precondition for every retention tactic that follows, and it is why the best retention strategies for Shopify all start with segmentation.

Retention marketing amplifies value; it does not create it

This is the sentence to remember from the whole hierarchy.

A loyalty program, a flow or a campaign takes the value that already exists in the brand (a product people want more of, a range worth exploring, customers with the intent to pay, an experience that felt good) and multiplies it by reminding, rewarding and recognising. When there is nothing to multiply, the same system produces open rates and nothing else.

That is why the first move when retention marketing is not working is diagnosis, not a new tool. It is the same principle the masterclass sets out under find the constraint: tactics are cheap, and the wrong tactic on the wrong constraint wastes a quarter.

A diagnosis table

Use this to work down the hierarchy. Start at the top; the first row where the signal matches is usually your constraint.

What the retention system should do once levels 1 to 4 are right

Once the product, the range, the customers and the experience are sound, the retention system has a clear job. It is not to give points and hope; points alone are a discount program in disguise. The job is to change a specific customer behaviour, and the chain looks like this:

Behaviour → incentive → action → value.

Start with the behaviour you want. If the constraint is the gap between the first and second order, design the program around that: a tier threshold set just above AOV so the second order unlocks something, a progress bar in the post-purchase emails, and a reward for order two that is a product or access rather than a percentage off. If AOV is low, tie greater rewards and access to a minimum spend. If frequency is low, look at the VIP tier setup: are the rewards exclusive enough, and do they reward buying within a time frame?

For customers who are already loyal, the behaviour to protect is staying. Do not discount to keep them; give them recognition, exclusivity, early access and events. That is the psychology that makes loyalty programs change behaviour rather than erode margin.

This is the model our Shopify customer retention tools are built around: a loyalty program that acts as a behavioural system on top of a brand where the first four levels already do their job. For how all the pieces fit together, start with the complete Shopify customer retention guide.

FAQ

Why is my retention marketing not working?

Most often because the constraint is above the retention system in the hierarchy: the product does not earn a reorder, the range has no natural second purchase, the customers were acquired on discounts, or the experience after purchase was poor. Email and loyalty amplify what is already there. Diagnose the five levels in order before changing the messaging.

What is the customer retention hierarchy?

The five factors that decide whether customers come back, ranked by importance: product, portfolio, acquisition quality, customer experience and the retention system (email, SMS, loyalty). The retention system is the most visible and the least important, which is why it is so often blamed for problems it cannot fix.

Should I pause my loyalty program if the product is the problem?

Pause promotion of the affected line, yes. Fix the product, then let the retention system amplify something customers actually want more of.