top of page
  • Writer's pictureAjay Row

The Math of Loyalty Programs - Part 5: Communication

Thanks for joining me here. This is the pen-ultimate article of the present series. Communication. Ideas. Uncreative creative. Data. How to get 70%+ open rates.


There are two contrasting communication requirements from a loyalty program: first, to strengthen the brand while building deeply personalised relationships with individual program members. And, second, to be the always-on, seedy underbelly of brand marketing, the shallows where unadvertised deals lie. Deep-discounts and the like, often portrayed as “member exclusives”.

There are several ways by which communication is delivered to members, broadly: 121 (email, mail, statements), digital and social (in which I include both general and personalised social media), point of sale communication (including staff training and incentives, critical but out of scope for this article), purchase or delivery and conventional media advertising. I understand that this list is limited and limiting, but it will do for this article. In fact, we will cover only a fraction of these opportunities while going through the key building blocks.

A quick backdrop. By signing on to a loyalty program, a member puts up a hand to say “Hey, I like you guys, I want to be your friend.” This is a privilege and an opportunity that marketers can either abuse or respect. Abuse, by spamming the member. Respect, by judiciously considering a key question before any communication is sent, “Why does this member want to get this piece of communication from us at this time and in this manner?”

Bumf answers, for instance, “Because our members love us and this is a fantastic deal, in fact we haven’t done one better in at least the last one, perhaps even two, weeks” or “I need to meet my numbers, I don’t care what it takes”, will eventually result in members treating communication from the program and brand as spam. And hence de-value both.

A propos, you know the old slaw about a 5% response rate being a great thing to achieve? Well that number is also a 95% non-response rate. Which means 95% of the folks with whom the program communicated have just been educated to ignore the program’s messaging. We really don’t want to do too much of that.

Let’s say you want to target a 60%+ rate of response; let’s figure out how to make this possible. There are three building blocks that build your campaigns response rates:

- First, marketing logic and IP. This basically is old-fashioned marketing thinking about the member database whereby the program managers determine who gets what communication when and how. A good idea is to also address the all too-often missing critical why? – see Para 3 and 4 above, they may be worth re-reading.

Two fundamental tools are segmentation (e.g. which customer is of potentially high value, who loves FB over email, who can be grown and who is likely to be lost) and cohorts (which member came in to the database when). Often loyalty folks create nursery programs for new members as they come in, a result of a nursery program is potentially high and low value members start becoming distinct. The point is communication to a HV high value member who came in just last week is different from a decade long low-value member.

Marketing logic is often reduced to composites (i.e. data markers or tags, they used to be called dummy variables in a simpler age) that basically use a combination of data, analytics and logic to “tell” the database how to communicate with whom and when. Today, you may well replace composites with some form of AI. But the wisdom behind composites remains, the intention to reduce complex operations into simple IF/THEN ideas, tasks easily executed by even us marketing folks.

A measure that the thinking is working is whether response rates rise or decline from campaign to campaign to a particular segment or cohort. If the former, Hallelujah, we’ve got it right. If that latter, you are in dangerous territory. Act before it all becomes critical.

- Second, database strategy. From a communication perspective, there are two things we need to worry about in our database strategy:

o Use the database most effectively – build those rates consistently and also ensure the database is not over-used. Finding that point of maximum effectiveness with no declines in response rates is part art and science. And perhaps you never find the ideal balance. No matter, for even looking for it is a good practice.

o How to keep building insight and operational capabilities on the database – the largest use of the program database is to communicate with members. If you cannot use it effectively to do so, and then consistently learn to do it better and still better, there is no point to the program. This is not to say that analytics and insights are not critical, just that this is the prime purpose. Communication is what moves members.

- Third, creative elements. The most important and enjoyable (unless you are an irrepressible data jock) part of the exercise is the creative part of our work. Here we have a few things to think about, only a couple are things our distant cousins in advertising would recognise:

o AB testing: no matter what you do, always test it against something else, create control groups and build learning about what works, and more critically what does not. A few tips that may help:

§ Swap-outs (suppressing unwanted names from campaigns) drive up response rates, initially there are far more folks not responding than responding, so there is far more data in the non-respondents to analyse, and hence learn from.

§ Use both universal and campaign-level control groups, this costs money, get that budget approved, it is worthwhile. (Not least because controls strongly justify our existence as direct marketers.)

§ Test the equivalent of subject lines (for opens), call-to-actions (for click-through-rates) and landing pages (for closes). Do this across all media, all communication, consistently, look for patterns, keep improving both good and bad odds.

§ Record notes of all learning in a manner clear to all. Review your notes periodically – you’ll be amazed at how much you forget. (Especially when you are at my age.)

o Templates: create templates that you store and reuse repeatedly, across cohorts and segments. If it ain’t broke don’t fix it, reuse. Keep looking for challengers, but respect those champions. They must have done something right! Some creative folks will hate you for this, but that’s ok, if you want love and work in loyalty get a uniquely understanding spouse. Or, a dog.

o Slots. What is key is to set a limit (ideally based on individual preferences but at the least based on segment preferences) to the number of 121 communications an individual will get from the brand / program per month. Usually 3 is about right, but this is a judgment call. The slots then rule database use, you cannot over-utilise it under any circumstances.

o Personalisation: is a good deal more than just calling a member by name. It is personalising the template to a particular member – recognize where he or she lives, who they are and much more. This works and is worth the extra effort. If your members are all getting the same material with just their names personalised, you may need to get back to the old drawing board. Point statements often have over 25 opportunities for unique personalisation, even while following segmentation and cohort logic.

o Operational effectiveness: create creative for media. This sounds self-evident but too many loyalty programs have confused a poster with an email, an FB post with a tweet, a display board with an ad. Make sure the creative “works” the way it should in that media. (This does not mean you cannot create templates, or cannot personalise effectively, but it makes the task a bit harder.)

o Properties: create “properties” i.e. regular communication members look forward to receiving and will open. For instance point statements – get your best deals into those, spend time designing them to ensure you get 70% opens and better. Create at least 2-3 properties besides the point statement, you cannot restrict yourself to that media alone.

o Ideas: most importantly perhaps, ideas. EVERY single creative you ever do must be driven by a strong marketing insight, an idea, communicated as effectively as possible to members. A creative without an idea, however clever it may be, won’t work as hard. Quantity in communication is rarely quality, which is why I can never understand why some of the world’s largest brands run their (usually bad) advertising repeatedly through a 20/20 match. Perhaps the brand gurus have a reason why, brand recall, TOMA, who knows what, but in loyalty I think this is unforgivable. You are looking to see terrific response rates. And data-driven insights and ideas are the way to get them.

o Awards and fun: a direct marketing and loyalty guy wants awards? Well yes. And why ever not? It’ll make your mom proud of you. But in good conscience pick ones you would like to have. An Echo? Yes. A Gold Pencil? Also, yes. Provided both had campaigns that worked super-hard and rightly deserved them. Make the creative fun to do. Our work is usually done in the corporate equivalent of dungeons and often in the dark reaches of the night. Most folks feel sorry for us. (Rightly so? Well we get to feel up data for fun.) I believe you owe it to yourself to have fun in creating great work you can be proud of. And enjoy the process of creation.

Always remember, as the advertising legend Rosser Reeves once said, “What do want from me? Fine writing? Pretty pictures? Or do you want to see the godammed sales curve start moving up?” In our field, the answer is clear. So learn the building blocks of 121 communication and apply these to your loyalty program. Drive solid business results that’ll make your CFO fond of you. And don’t forget to have fun.

Related Posts

See All


bottom of page