Trust Me, I'm a Marketer: Behaviour

The second of the “Trust Me, I'm a Marketer” pieces, from around 2010 - and the one I'm fondest of in hindsight. Read past the segment mechanics to the warning near the end. The argument that leaning too hard on your best customers eventually burns them out is, in embryo, the idea I now build everything around: a responsive, trusting audience is an asset you can deplete. Fifteen years on, the principle is the same; only the precision has changed.

For effective database marketing we need to know what our database can realistically deliver in sales. An event rarely sells out entirely to people already in the system. When I was studying for my Master's, a guest speaker - the marketing manager of a regional arts centre - described a challenge his team had been set: sell a 1,200-seat contemporary dance performance to capacity, marketing only to existing customers on the database. They got creative, and they did it. But, he said, it wasn't a sustainable way to market. I agree. So we need to know what the database will normally deliver - a simple, little-used trick.

The simplest method: look at a comparable past event and ask “how many of the people who came had been before?” On average you'll find 40–60% of an event's customers had not been before. So the database can only be expected to deliver what's left. For simplicity, call it a 50/50 split.

Now you can size the job:

• You need: 1,000 tickets
• People buy, on average: 2 tickets per booker
• So you need: 500 bookers
• If 50% come from the database: 250 bookers from the database

Einstein defined insanity as doing the same thing over and over and expecting different results. The same is true of database marketing: repeat the same activity and you'll get the same outcome. At least now we're armed with two things - what the database can deliver per show, and which segments have the highest propensity to re-attend. I'm not suggesting you chase 90% of an event's sales from the database, but you do need to know the actual impact of your direct marketing. Only then can you make and measure changes.

The four cornerstones

Over my time as a database marketer I came to think of database marketing as resting on four cornerstones:

• Segmentation
• Interest
• Motivation
• Behaviour

To be good at it, you need to be accounting for at least one of them. The first piece covered segmentation; here we look at behaviour.

Behaviour

A quick recap: Recency and Frequency determine propensity to attend - how likely someone is to come back in a given period. We categorised customers in a matrix; here I've borrowed the Boston Matrix's language to name them:

Each behaves distinctly - worth understanding before you communicate. It also matters to know when a customer has hit their attendance threshold. If someone attends on the same date each year, has done for five years, and has already booked next year's visit, it's clearly wrong to email them weekly about a show that isn't selling. You risk putting them off entirely, because the only contact they have with you between visits is a marketing inbox filling up.

Cash Cows

What we know: they've attended recently and are comparatively frequent. A little analysis shows they account for roughly 75% of revenue each month. They are also more likely to:

• Book further in advance
• Come in smaller groups
• Re-attend more than twice a year

What we can assume: they know us - where to park, where the bar is, what time to arrive. They probably come because they like coming, so they're less tied to a single genre. Sell them only what they've seen before and they may go elsewhere for the rest, so be comfortable marketing different work to them. If you produce a brochure, these are the people who should get it - and arguably no one else.

Rising Stars

Attended once, recently - most likely their first visit. The challenge is the all-important second purchase; typically only 1 in 25–50 come back within 18 months. They account for about 10% of monthly revenue and tend to:

• Book last-minute, close to the day
• Come in larger groups
• Buy more expensive tickets (a good seat for a good time)
• If they return, wait 6–12 months before doing so

What we can assume: because they've been only once, it was probably the event that drew them, not us. So here we should offer them the same kind of product as last time.

Question Marks

The interesting bunch: been a couple of times, but not for a while. Why not? They were probably once Cash Cows who slipped, unnoticed, into the Question Mark group. Given it takes 25–50 Rising Stars to make a Cash Cow, losing one this way is costly. They're around 10% of monthly revenue and share traits of both:

• Mid-range booking time (not last-minute, not far ahead)
• Smaller groups
• If re-engaged, more likely to return within 6 months

What we can assume: if the artistic programme changed, they may be voting with their feet. If it hasn't, the reasons are many - busy, a change of life stage, a new family. One thing is certain: they used to come, and used to like us. So sell them the benefits of going out, and of coming back to us.

Dogs

Usually only about 5% of the monthly take, but probably the largest segment on the database - lots of people who haven't been for a long time, few of whom return. When they do book, they behave like Rising Stars:

• Their booking time is close to the event
• Larger groups
• More expensive tickets

So when you need your 250 database bookers: about 75% will be Cash Cows, who aren't product-specific and won't thank you for a single-event push - they're booking off your regular newsletter. Because they book further ahead, as the event nears and isn't selling that segment yields less, and you turn to Rising Stars and Question Marks. Still struggling? Then the Dogs. And if you're relying on Dogs to finish the job, the product was probably a curve ball for your Cash Cow market - you're really finding a new audience for it, which you would have known before you started.

A warning

There's a danger here. We've seen clients realise how strongly Recent-and-Frequent customers respond - and lean on them. But sustained, focused reliance on that group leads to frustrated customers, bombarded with activity, and ultimately declining response. You get a flood from Cash Cow to Question Mark: fewer Cash Cows to market to, and those that remain won't thank you for the constant contact.

In summary

It's vital to spread the marketing load proportionally across the four segments, and to put in place a plan to keep replacing the customers who graduate into - and eventually out of - the best-customer group. Keeping that balance is the problem the rest of this work has been about ever since. It's the thread that runs straight into what N.E.R.D builds today - only now the balancing act happens patron by patron, not segment by segment.