Search
SUBSCRIBE
WEEKLY STRATEGY
FREE AUDIT
THE SOFTWARE
calendar-dots
LET’S TALK
Apr 29, 2026
Not All Customers Are Created Equal
Not All Customers Are Created Equal
00:00
09:13
Transcript
0:00
Last week we talked about benchmarks: using external data to get an honest read on where your business actually stands relative to your peers.
0:08
This week, I want to go inside the customer file and look at something that most brands know is true in theory, but almost never act on in practice.
0:17
The customers in your database are not equal, and treating them as if they are is one of the most consistent and expensive mistakes I see in DTC.
0:27
[upbeat music] Hi, I'm Alex Orly, and this is Hard Margins, your weekly e-commerce brief, brought to you by RetentionX.
0:40
[upbeat music] RetentionX is the only integrated growth platform built for Shopify businesses. One clean source of truth for your customer data and the tools to act on it, no data team required.
0:57
Today we're going into RFM: recency, frequency, and monetary value. The simplest and most practical ways to turn your customer list into something that actually informs decisions.
1:08
By the end of this episode, you'll have a clear framework for segmenting your file and a concrete sense of how to act differently on each segment.
1:16
Inside Shopify, a one-time discount shopper and a multi-year VIP show up under the same label: customer. That's not just an oversimplification.
1:26
Operationally, it creates real problems because every decision that gets made about customers ends up applying equally to people who are genuinely valuable and people who are actively diluting your economics.
1:39
In almost every seven, eight, or nine-figure brand that I've seen, the customer file follows the same distribution.
1:45
A small group, often somewhere between one and ten percent of the total file, drives a truly disproportionate share of revenue and profit.
1:54
There's a healthy middle that contributes stable volume but is not critical, and then there's a very long tail of one-time, low-value, heavily discounted buyers who generate a lot of noise and very little margin once you subtract COGS and discounts.
2:10
By treating all of these groups the same in your CRM, your paid media, and your email program, you end up overspending on dead or low-value segments, underinvesting in the people who are actually funding the PNL, and operating without any clear sense of who deserves what.
2:27
RFM is the cleanest way to fix that. RFM gives every customer a numerical score, a quartile score on three dimensions. Recency is how recently they've purchased.
2:37
A high recency score means they're active in the relationship. A low score means they're drifting or already gone. Frequency is how many orders they've placed. High frequency suggests habit and trust.
2:50
Low frequency suggests a trial buyer or someone who came in once and never engaged again. And monetary is how much they've spent in total. High monetary represents a meaningful contribution to revenue.
3:02
Low monetary is largely noise once you account for COGS and discounts. The practical implementation doesn't need to be complex.
3:10
A simple one-to-four scale on each dimension is usually more than enough to surface the meaningful separation in your file.
3:17
Once you have RFM scores, you can group customers into segments that carry real commercial meaning. Your top customers are recent, frequent, and high spend.
3:27
These are the customers who drive contribution margin and fund everything else: new product tests, R&D, acquisition experiments, brand investment.
3:37
They deserve disproportionate attention and highly personalized treatment. High potentials are recent, have placed a few orders, and show decent spend.
3:46
These are your next generation of top customers if you engage them correctly.
3:51
The leverage here is product education, smart cross-selling, and really anything that increases frequency and monetary value before the relationship plateaus.
4:05
The loyal middle are regular customers with solid but not outsized spend.
4:10
They're reliable, profitable, and they deserve consistent, respectful communication, but they don't need the same level of investment as, say, your top two tiers.
4:19
Bargain hunters or promo-only buyers tend to be frequent but low spend, and their engagement is almost entirely discount driven.
4:27
They can contribute volume in controlled contexts, but they're genuinely dangerous if they start to dominate your file because they condition the database to expect discounts and erode the economics for everyone else.
4:39
At-risk or fading customers have good frequency and monetary scores, but their recency has dropped. This is probably your highest leverage retention target.
4:48
They've demonstrated real historical value and a real relationship, and the window to recover them is probably closing. And then there are low-value one-timers: low recency, low frequency, low monetary.
5:00
This is the group most brands spend far too much time and money trying to reactivate.
5:05
In most cases, the honest answer is a light touch sunset sequence and then a reallocation of that budget towards segments that actually have upside. Much better use of your capital.
5:15
When you overlay RFM with concentration data, the picture comes into focus pretty quickly.
5:20
The top customer segment, often just, again, one to ten percent of the total file, typically contributes somewhere between thirty and sixty percent of revenue.
5:29
Add in high potentials in the loyal middle, and you're looking at your top twenty percent of customers generating the vast majority of value.
5:37
The entire long tail of one-timers, bargain hunters, and low M-score buyers contribute surprisingly little profit once the disco- once the discounts and the COGS come out.
5:48
The practical implications of this are uncomfortable but important. Losing a single top customer can hurt more than losing a hundred one-timers.
5:57
Directing retention efforts towards low-value segments is often a genuine misallocation.And over-discounting to keep bargain hunters technically active can actively damage the economics of the segments that are actually funding the business.
6:11
Once you have a working RFM segmentation, it functions as a routing layer for money, attention, and effort across every channel.
6:19
In email and SMS, top customers should receive fewer, far more personal, and more value-dense communication.
6:26
That's early access, exclusive drops, things that reinforce their relationship with the brand rather than training them to wait for a discount.
6:35
High potentials should get onboarding style content, product education, smart cross-sells designed to increase their frequency and spend.
6:44
Low-value one-timers were into light touch or a deliberate sunset, not the same budget and effort as everyone else.
6:52
In paid, the implication is that lookalike audiences should be built from your top and high-potential segments, not from all purchasers. The channel learns from whoever you give it as a seed.
7:03
If you give it a full file, including every bargain hunter and one-timer, you're teaching the algorithm to find more of those.
7:10
Low monetary segments should generally be excluded from retargeting where it's cost-effective to do so.
7:15
In CX and operations, top customers should be flagged in your support system with differentiated SLAs and escalation rules.
7:23
At-risk customers with high historical RFM scores should be treated as high-priority save cases, not generic reactivation tickets.
7:31
And in product and promotions, the products that over-index in your top and high-potential segments deserve more real estate and ads and emails, while products that primarily attract low monetary discount-only buyers should be used carefully or deprioritized.
7:46
If I were sitting in your seat, I'd ask for three things from the team this week. First, a current RFM segmentation of the entire customer file. How many customers sit in each key segment?
7:57
What does that distribution actually look like? Second, revenue and contribution margin by segment. What percentage of revenue in CM1 comes from the top two tiers versus the rest?
8:08
And how much are you currently spending on each group in media, discounts, and customer support efforts? And third, one concrete action per high-value segment.
8:17
What specifically are you going to do differently for top customers? How are you going to accelerate high potentials? And where are you going to stop over-investing in low-value groups? RFM is not a reporting exercise.
8:32
It's a lens that changes how you allocate almost everything: budget, attention, creative, customer experience, and support.
8:41
Once you've seen your file through it, it becomes very hard to go back to treating everyone the same.
8:46
Next week, we'll connect this to cohorts, so you're not just seeing who matters today, but whether the new customers coming into your file are on track to become the kind of customers you actually want more of.
8:56
[upbeat music] Thanks for listening. I'm Alex Orley. This has been Hard Margins, your weekly e-commerce brief, brought to you by RetentionX. I'll see you next week. [upbeat music]
Hard Margins
Recent episodes
Benchmarking for DTC Growth
Apr 29, 2026
The LTV Playbook
Apr 29, 2026
Scaling to $100M on Shopify
Apr 29, 2026
