Customer segmentation “best practices” can lead to analysis paralysis. Here is a simple segmentation strategy that can work for every brand.
This content was originally published in the No Best Practices newsletter on 10.30.2022.
Over the course of my 10+ years working in eCom, I’ve worked with more than 20 different brands across a variety of product categories. Most of these brands had one of two approaches to customer segmentation:
- A complex model with 6-10 (or more!) segments roughly based on recency, frequency and monetary value criteria.
- No segmentation at all; everyone gets the same message at the same frequency.
Is there a significant difference in performance between these two approaches? Not really! And that’s because brands who take the first approach rarely do anything meaningful with all those segments. At most, they contact certain segments more frequently than others.
So, is segmentation a complete waste of time? Not if you do it the right way. I’m going to explain why so many brands get segmentation so wrong and outline the only segmentation framework you really need to start making some sweet, sweet incremental moolah.
The Customer Segmentation Trap
Certain SaaS companies have crafted a “customer journey” narrative, in which there is some magic formula of marketing touches and offers that can unlock repeat purchases from every one of your customers.
This is a load of bull. If you are a mono-brand eCom retailer, especially if you sell a non-consumable product, you’ll be lucky if 35 of every 100 new customers you acquire EVER returns to shop from you again.
But these Saas companies will encourage you to collect as much data as possible, and slice up your customer base in a thousand different ways in an effort to unlocking the secret of ~customer lifetime value~.
Never mind the fact that most brands struggle to develop enough creative assets to send out five emails per week. These SaaS hucksters want you to develop 5 different email versions per day.
Of course, these software companies never talk about how to execute on their carefully crafted segmentation. That’s the client’s problem. But brands have dropped huge cash on an annual agreement, so they have to do something.
Most marketers do what they have the resources and the mandate to do, which is increase message cadence and test different promo codes.
Are these tactics driving incremental revenue for brands? It’s hard to tell, because most brands measure these tactics on a last click basis, so incremental revenue is never part of the discussion.
That’s why most segmentation (like most retention marketing) is either value-neutral or value-negative. But “best practices” segmentation does serve one purpose: it creates busywork that enables empire-building within “enterprise” organizations. I guess that’s a win for toothless middle-managers everywhere. But that’s not you, because you read this newsletter.
Should I Segment At All? A 4-Point Checklist
Segmentation can help you grow your brand. But you need to check off a few criteria first:
1. You have a meaningful repeat customer base. Segmentation helps you to generate incremental revenue from existing customers. But for the benefits to outweigh the effort, you need to be generating meaningful revenue from existing customers.
If your business is less than a year old, less than 25% of your revenue comes from repeat customers, or your active repeat customer base is fewer than 5,000 people, focus on scaling acquisition instead.
2. You have studied your customer & baseline marketing is locked in. You need to be able to say “We are speaking a certain way to our customer base because our customers value these things”. You need some kind of default marketing strategy that is not simply pointing various sources of paid traffic at your website and crossing your fingers, or drafting off decades of built-up brand awareness.
When you segment, you are making the choice to diverge from your default way of speaking to customers for some meaningful reason. If you don’t understand your customer base, you can’t do this effectively.
3. You have the resources and willingness to version creative. I’ll repeat myself because this is so important: segmentation = the choice to speak to a certain group of customers differently. If you’re not comfortable with this because it’s not “on brand”, or because your creative team is backed up creating 3 promo emails per day, then segmentation is not for you.
4. You have the ability to study results at an audience level. The benefits of good segmentation are rarely revealed through last-click analytics. You need to build holdout tests. This can be done with a software tool or good old fashioned Excel/SQL/Python analysis of raw data.
Foolproof Customer Segmentation
Ok, you’ve determined that you have what it takes to slice and dice that customer file effectively. Now it’s time for the hard part–where to begin?
You want to split your customer file into big, causative buckets. You want to be able to say: this is a definitive moment where additional lifetime value is won or lost. We are going to do something to help mitigate that loss.
This is a big part of the retention audits I do for my clients, because many of these critical moments are unique to a product category and to each individual business. But there is one pattern that I have seen time and time again, that any brand can leverage to drive a lot of value. If you’re ready to implement meaningful customer segmentation, this is where you should start.
The customer lifecycle has three phases: prospect -> nurture -> loyalty. Prospects have zero lifetime purchases with you; this is the domain of acquisition marketing. But, just because a customer converted once, it doesn’t mean that they’re completely bought into your brand. This is why only 20 to 30 of every 100 new customers you acquire EVER comes back. I wrote more about this here.
If you’re looking for the bottleneck to your retained customer growth, this is it. To make a customer loyal, you have to get them to come back to you. And most don’t. In fact, most customers do not become “loyal” until the 3rd to 5th purchase.
So here are your two segments:
Prospects & Customers With 1-3 Lifetime Purchases: These customers are still warming up to your brand. You need to assume a low level of context, and below-average levels of trust.
Feature your best selling products and explain why they’re great. Reinforce trust with social proof. Educate the customer about your brand–why you exist, why you’re different, and all of the products/categories you offer.
Keep it light on the email frequency unless these customers are clicking on every single email that you send. Otherwise, you’ll alienate them before you have a chance to build a rapport.
Your goal here is to move each customer on to a subsequent purchase before they lapse. There are key timeframes when that is more likely to happen, which are different for every brand. Once you’ve exited those time frames, it’s highly unlikely these customers will engage with you again.
Customers With 4+ Lifetime Purchases: These customers are totally bought-in to your brand. You can assume a high level of context and high levels of trust (don’t mess that up, though).
These are the customers who are most likely to care about new product launches and less popular categories within your assortment. But, because they’re more likely to buy anything, it’s going to be harder to win an incremental sale or incremental spend.
You can contact these customers more frequently, but pay attention to open/click rates so you don’t piss them off and lose them.
Bonus Third Segment–Disengaged Customers: These customers have stopped opening your emails and stopped visiting your site. You should set up some recurring programs to try reengaging them. But at a certain point you have to give up and admit to yourself that continuing to market to these folks is not a high-leverage use of your time or money.
Channels Vs. Experiences
The goal of segmentation is to change customer behavior to your advantage. You are up against human nature, in some cases. This process takes TIME. One email is not going to move the needle.
You need to think about the entire experience your audience receives, across your marketing mix. If you send three emails a day, every day, changing one subject line is not going to have a meaningful impact.
Instead, you should leverage these segments across your retention marketing channels and make sure that the total experience–merchandise, copy, content, creative, offers–meets that segment where they are.
This doesn’t have to be hard, and it doesn’t have to mean doubling or tripling your creative output. I’ll go into detail about building effective creative teams and processes in another newsletter if that’s something you’d be interested in–just reply and let me know.
Here are a few things to consider:
Frequency: If you email your prospects and first-time buyers every day, the good ones will leave. You will wind up with an email file full of promo-hungry vultures. And that can be very profitable, but you’re leaving money on the table.
Marketing Messages: We often talk about new products or features extensively within our brands and teams. But your prospects and new customers haven’t been in the room with you for the past six months; you need to step back and explain why these things matter to them. If you can’t do this, your product strategy may be the problem.
Similarly, loyalty messaging is going to fall flat with this group. Pushing loyalty programs on new customers is bad strategy. It’s like proposing on the first date.
Merchandise: Different segments of your assortment appeal to different customers. The broader your assortment is, and the more frequently you refresh your assortment, the more true this is. It’s so business-specific that I can’t give you any meaningful advice here. But figure it out, or hire me to figure it out for you.
Offers: Prospects and new customers are not going to buy into some complicated points scheme or “spend more, save more” tired offer. You’re essentially saying “jump through hoops for this thing you barely know about, much less like”.