Are You Doing “segmentation” or Segmentation?
For purposes of this post all versions of “segmentation” that include air quotes” ” and a lower case ‘s’ are not segmentation at all.
Sadly, most versions of claimed Segmentation are really little more than the impersonating cousin named “segmentation”.
None of the following is real Segmentation;
- Mailing the same thing to different audiences. An example is the agency that rents Humanitarian lists and Social service lists and sends both lists the same thing.
- Mailing different packages to the same people. An example is dividing your house file or your final, merged acquisition list into two random nth groups then sending different mailings to the two groups. Hint: if you create random nths you’re assuming everyone is the same. Sending a “we help seniors” mailing to one of your random nth’s and a “we help kids” mailing to another of your random nths is equivalent to sending the same thing to everyone. Finding out that the seniors message beats kids message is finding the best, weak average. Do you think it was preferred by everyone? Of course not, which is why this is “segmentation”.
- Breaking out results after the fact using different attributes. If you utilized either of the two approaches above and then prepared reports showing different outcomes by RFM bucket, other internal labels, or using 3rd party data then you are still stuck with ‘segmentation’. No number of groups –even if broken out eight ways to Sunday– will fix your starting point problem.
Here’s a definition of and process for Real Segmentation.
Creating different groups of people based on differing needs and messaging them differently because you believe the juice will be worth the squeeze.
If your segmentation mostly lives in PowerPoint (e.g. clusters, personas) or a spreadsheet then there is a 99.9% chance it’s a “segmentation”. If it lives in a process that impacts every step of your fundraising it’s at least on it’s way to a Segmentation.
To avoid any wiggle room, here’s a simple, concise set of steps. All steps must be followed to do real Segmentation.
- Step One. Define two groups, let’s call them A and B, and write out detailed guidance on how fundraising imagery and copy should differ between the two. You’ll want a strong reason to believe group A is different from B and that those differences explain their giving. This last part is the juice-being-worth-the-squeeze requirement. If I say Group A gives ‘online’ and Group B gives “offline” then I’ve missed the mark. That describes the way they give, not the reason for it. Same goes for all our internal labels – sustainer, one-time, etc. And lets return to our senior vs. kid mailings. Do you think some people prefer to help Seniors and other, different people prefer to help kids? Maybe but why? What makes those people different in that way?
- Step Two: Tag individual records in your CRM or acquisition mail merge list as belonging to either group A or B. Hint: these groups are mutually exclusive and if you have to make arbitrary choices on whether to put them in A or B then go back to Step One. If you don’t have any idea how to create groups A and B that will be reliable, go back to Step One. You may still have some portion of your list tagged as ‘unknown’, which is fine.
- Step Three: Create two offers, A and B.
- Step Four: Offer A only goes to Audience A and Offer B to Audience B. If you feel the need to send Offer A to Group B then you probably didn’t do a good enough job on Step One. However, if you want to do this to affirm your targeting and message then do it in digital first, this can be done for hundreds of dollars in a week, not tens of thousands over months.
Your aim with all this is two controls, one for Audience A, another for B. The path to growth from this starting point is the traditional, incremental testing with each group to beat the control , but more importantly, it’s repeating steps 1-4 within Audience A and separately, within Audience B. You don’t have one audience and you almost certainly have more than two.
Do you think all people who prefer the kid (senior) mailing are the same? Don’t we think there is a way to frame the kid issue differently that might really match how some people in Group A think but not all of them?
The real growth path is segmenting based on who your audience is independent of you in the picture.
This is about the old ‘start from where they are’ trope. Your donors do not intrinsically care about your charity nor your beneficiaries. Where they are is pre-baked and often, hard-wired. If you match your pitch to their hard-wiring as opposed to your organization’s brand or issue A, B or C then you’ve created a true connection and the donor’s intrinsic self will be more likely to care.
You’ve probably heard of needs-based segmentation, it typically lives in a PowerPoint making it a “segmentation”. Needs are real but they are deeper than a need to help kids or seniors or your charity.
I am a parent who skews Conscientious and prefers a loyalty frame and using attentiveness as the emotion. I’ll help kids if the message is framed to match who I am. My neighbor has kids the exact same age. He skews Agreeable and prefers a caring frame using Compassion as the emotion. We’re the same age, gender and race and have both given to the same type of charities before.
In a “segmentation” we get the same mail tailored to neither one of us but maybe, just maybe, one of us has a spare minute and gives it to you to put in the necessary mental energy required to find himself in your letter and give.
It’s two different mailings in a real Segmentation. If our mail gets switched neither of us responds. If it doesn’t, we both do.
Kevin
5 responses to “Are You Doing “segmentation” or Segmentation?”
Behavioral Science Q & A
Integrating an individual giving appeal with other communications from a charity can have both positive and negative effects, and the outcome largely depends on how it’s executed. Advantages of Integration Brand Consistency: Maintaining a consistent appearance and messaging across all communications can reinforce the org’s brand identity and strengthen brand recognition and trust among your […]
Read Full Answer
I’m not aware of any in-market tests specifically comparing recurring vs. gift frequency language. I suspect the answer might not be the same with all gift frequencies, nor with all people. It sounds like a great opportunity for you to test and find out what works for your audience. Based on the literature, here’s a couple […]
Read Full Answer
Based on what we know from existing data, those renewal notices can actually be pretty effective in getting people to donate. They tap into our psychology – creating a sense of urgency, reminding us of past support, and using personalization to make the message hit home. They’re playing on our natural tendencies to feel obligated […]
Read Full Answer
Interesting question. I had a quick look at the testing done on this topic. On the positive side, in all cases, over half of donors decide to cover the fee. In some cases, it goes as high as 65%. Not a negligible percentage at all. Here’s another test from iRaiser showing consistent results (see point […]
Read Full Answer
There’s just one thing to consider when designing a supporter journey: the supporter. More specifically, you need to take into account: Who the supporter is i.e. their identity, which is the reason they support this cause, and their personality, which describes the way they “see” and process the world. These will determine the kind of […]
Read Full Answer
I’m not an expert in this but a quick search surfaced this article on the effect of tax reforms on 2019’s charitable giving. The researchers didn’t find a reduction. Actually, they observed an “increase in charitable contributions in 2019, even with the lower tax rates and the dramatically smaller number of taxpayers who itemize their […]
Read Full Answer
Definitely agree that getting to ever more focused forms of segmentation make sense, so the thrust of this is right on.
However, I’d look at this as more of a spectrum of segmentation than You Must Be At Least This Segment-y to Ride This Ride.
Is mailing the same thing to different audiences ideal segmentation? No. But is it segmentation or Segmentation or SEGMENTATION with holiday lights on it? Certainly – it’s segmentation of in versus out.
Because if DonorVoice taught me nothing else, it’s that *sending nothing is always a choice*, that it should only be done when it improves on silence, and that happens less frequently than we think it does.
In that way, when you get a UVA alumni mailing and your neighbor gets a mailing for his particular house of worship, neither of you got the other one’s mail piece, nor should you have. And if you switched mail, as in your example, neither of you would respond to the other one’s piece.
I think of the classic DV example of how segmentation based on knowledge of the organization for Amnesty International increased retention by sending cultivation pieces to only those with low knowledge. That’s in/out segmentation, it’s the bedrock for audience selection, and it’s Segmentation even if all the people who got the piece get the same piece.
That’s one of the spectrum. On the other end of the spectrum, we shouldn’t pretend that A/B isn’t the platonic ideal of segmentation. Because if DonorVoice taught me a second thing, it’s that we all have multiple intersecting identities. So the ideal segmentation and customization is similarly multiplied. It’s customizing the copy to whether someone is “lapsed”, what area of the mission they want to support, what language it should be, whether they have the disease themselves or whether they care about someone who does, what their level of satisfaction with the organization is, whether they want support to go local or international, what their professional association to the mission is, what their propensity is to make a monthly gift is, what their financial situation is, what channels they prefer to communicate through, where they live on the Big 5, etc. etc. etc.
The goal is that level of one-to-one communication (perhaps through some sort of constituent data platform? :-)) across channels.
And every step toward that ideal is a valuable one. Sending full file now? Try an in/out segmentation. Only segmentation on transactional data? See if you can add reason for giving both to the in/out segmentation and customized communication. Have your first type of message change done? It’s time for your next one.
So I’d quibble on staking out a position at a point on this spectrum and saying “this is true segmentation” because it doesn’t just exclude those things that are worse; it could limit vision to what’s better.
Mr. Ellinger, always a pleasure my friend and as always, smart, useful commentary. I think we’re in more agreement than not but let me counter your quibbles with quibbles on your quibbles. Double-quibbling.
1) In-market vs. out is really about selection vs. any form of segmentation. UVA doesn’t have a problem mailing non-alums nor do the Lutherans struggle with mailing too many atheists or pagans. The UVA problem is a lousy, one-size fits all approach to their alum fundraising. Lousy as evidenced by the plummeting participation rates. I don’t think this is much of a sector problem, at least on a relative basis. Progressive groups know how to avoid conservative group lists and vice versa and a myriad of permutations.
2) Choosing from your ‘in-market’ to mail fewer folks is about efficiency, not segmentation, which should be about effectiveness.
3) Selection and efficiency aren’t in place of or substitutes for segmentation, they are table-stakes precursors. Neither should be thought of as a way-station to get beyond one-size-fits nobody, which is exactly where any group is that uses selection or efficiency as “segmentation”.
It seems we’re defining segmentation downward.
Anecdotally, I get nothing but head nodding when asking folks if people are different. The heads keep nodding when asking if most fundraising ignores this.
There is no version of selection or efficiency in your examples that addresses the final list of folks getting solicited being sent the exact same thing. A real segmentation is looking at the universe you will solicit, after your selection and efficiency efforts, and figuring out how to not treat them the exact same.
Good to quibble with you once again – I always come away smarter.
Let’s take a thought experiment on the lower end of the segmentation spectrum. Through some means — trickster god, cursed amelet, ill-adviced wish on a Zoltar machine, etc. — you can only send the same communications to donors next year. Could you still do big-S Segmentation?
I would say of course. Without customization to think of, you’d have to spend the full year coming up with better and better ways to get communication X to the right people. You’d take a full file mailing, improve it to last 24 months, then go to RFM, then add in identity, then loyalty and satisfaction and identities 2 through 9 and demographic data and when-they-last-came-to-the-website and are-they-volunteers and so on and on. All of this would create a rich data stew that could get you a great in/out Segmentation.
It would absolutely chaff that Loki/Zoltar/amulet made it so you had cut potential valuable donors because this wasn’t the communication for them. And it would hurt that you couldn’t create the ideal communication for the one who got the communication but could have received a better one.
But is it Segmentation and improving the donor experience within the confines of having one hand tied behind your back? Absolutely. Efficiency shouldn’t be a curse word because, from a donor’s perspective, it means not wasting their time.
On the other side of the spectrum, I too don’t want to define down segmentation and that’s why I don’t want to limit it to an A/B world. We should not rest until every person gets the communications most likely to make them give, where those communications are different from each and every other person.
We could without breaking a sweat list 20 variables that matter to a donor segmentation. With a little sweat, I think we could get to 30. Now let’s say there are only two versions of each of these variables (a problematic exercise for many variables, but it makes the math easier). That’s 2^30 possibilities, for over a billion different versions. More than enough to make every American’s journey unique (and if you add in the rest of the world, just assume “country of origin” as another variable).
In other words, if you look at three questions:
– Who should get communications?
– What communications should they get?
– How many variables should we use to determine this?
You’ve correctly identified that the best (but not only) segmentations will involve both of the first two. But I don’t believe that “one” is the correct answer for the last question. Our journey should definitely go through where you’ve identified as Segmentation, but it neither begins nor ends there.
That’s why I’m willing to celebrate “maybe we shouldn’t mail the full file every time” and “let’s add customization #27 for whether they have kids at home or not to our digital printing” — each is a step forward. And I think we can get to the one-to-one marketing revolution through evolution, with each step proving its value along the way and some bold pioneers taking leaps where others take steps.
So in short, I don’t want something to have to be perfect to be labeled segmentation. And I don’t want something to be labeled perfect that is in fact imperfectable.
[…] disagreement, as reflected in their comments, is worthy of further exploration. And since the original post was cut short by Christmas we’re […]
[…] Kevin noted in Are You Doing “segmentation” or Segmentation? current practices –although high-tech and quite sophisticated – are still little more than […]