Loyalty Programs

Right now there seem to be a few people that I’m aware of who write about the online poker industry from an insider’s perspective and aren’t shackled with having to take on a corporate voice for their employer. One is Yair Panet who was formerly at iPoker and another is Kim Lund who was at BWin. Yair has written a couple of guest posts here on Bill’s Poker Blog titled, Poker and Social Media and Finally, something original!. Yair has also written as a guest blogger on Kim’s site and his most recent post titled, Show me the money? Loyalty plans, rakeback and everything in between, seems to have stirred up a lot of discussion about loyalty programs that has even drawn in Jonas Ödman from Bodog as well as people from PokerStrategy and consulting firm HUM4N.

First off, go read the post. It’s excellent and the follow on comments are really interesting.

Are you done reading it? Okay, let’s proceed.

I wanted to write this as a comment on Kim’s blog but as you’ll see I’m far too verbose for a simple comment 🙂

First off, it’s always been one of my pet peeves when people call whatever they’re offering to players a “loyalty program.” They aren’t loyalty programs. Here are a few common definitions of the word loyalty:

steadfast in allegiance or duty
firm: unwavering in devotion to friend or vow or cause

I’m sorry but few if any so-called loyalty programs offered by online poker rooms meet that definition. I looked far and wide and I never saw a definition for loyalty which read, reward people based on activity regardless of whether they are any more or less likely to engage in future activities, which seems to be the cornerstone of nearly all loyalty programs offered within the industry.

In the comments on Yair’s article Denis Campbell hits the nail firmly on the head when he says:

Loyalty systems based on points, bonuses, rakeback etc are nothing but a deferred price reduction. One thing one knows about customers who choose on price is that their only loyalty is to the last deal.

That is the complete antithesis of loyalty. And any system based on that model is not a loyalty program in the strictest definition of the term.

I think most of the people commenting agreed that the vast majority of the industry is simply copying what everyone else is doing. That has been true for affiliate models, loyalty programs, promotions, signup incentives, and just about everything else. I am actually quite disgusted at the total lack of innovation and creativity in the poker industry today.

This is why I so strongly disagree with Jonas Ödman and Patrick Selin (both from Bodog) when they say that there will be no consolidation. There are simply way too many operators who have no idea what they’re doing and are adding no value to the marketplace for them all to be able to survive.

So what is the answer? Well, Kim starts to get at it a bit:

You HAVE to start this entire discussion with the incorrect valuation of players that the current rake/points model creates in mind. Otherwise you end up drawing good conclusions from bad premises.

“that the winning players are the ones who will get the most rewards”

To be be correct it is the active players that get the most rewards like Jonas points out. These correlate quite well with the winning players but not entirely.
A model based on activity is the natural way to go about it, it just doesn’t work very well for poker when one player’s activity over another’s depends so much on luck.

“. Simply because you don’t call it rakeback but loyalty – doesn’t mean you suddenly solved the rakeback war problems.”

Exactly.

First off, Kim makes a good point that you have to start off any discussion with the assumption that the industry has no clue about how to value a player. Kim, Yair and I have had some good debates and I respect their insights but because they come from a network model I strongly disagree with them on several issues because I feel that their opinions are shaped at killing the rakeback monster and making sure skins aren’t poaching from each other. Well, that’s all fine and dandy and I can totally appreciate why this is a concern for them but why should your customers give a hoot whether or not you have those problems? That is why I’ve frequently said the entire network model is flawed. There’s too much emphasis on keeping the skins in line and monkeying around with trying to achieve some sort of equality in the poker ecosystem and not enough on what’s best for the player. Sure, the network can say what it’s doing is best for the player but if the player doesn’t see it that way then . . . bye bye. Your operational constraints should not be a problem you push onto the users.

In my opinion, the traditional view that players can be looked at through a simple lens of net winner or net depositor is equally as flawed as ranking players on rake generated. What is the net value of $1 deposited onto a poker site? How many times does that dollar get churned? How many hands does it pass through and how much rake does it generate along the way? Do you have to ply the guy with bonuses to get him playing or is he a steady depositor? Does he dump it all in one hand or does he play for several hours on that $1? Is $1 wagered in a high-stakes game the same as $1 wagered in a micro-stakes game? These are all questions that need to be accounted for when putting a value on $1 being deposited onto a poker site.

But the bigger issue here is that one size does not fit all. Several people made the frequent flyer miles comparison but that’s not a great analogy because it too is a one size fits all model. You fly X number of miles and you get X (or sometimes Y) number of points. You can cash those points in for upgrades and free flights, blah, blah, blah. Booooooooring! It’s a copycat solution in a copycat industry.

In reality you have a strata of people who crave exactly that model. Your grinders want to be paid Y for doing X. So do it. I can think of at least half a dozen programs you can come up with that would give them exactly what they want and still incorporate loyalty. For instance, 40% rakeback and a one-time 10% bonus on top of their 40% if they generate X number of dollars in rake per month every month over a certain period. Sort of like Full Tilt’s IronMan except for rakeback. And if you’re so bent on controlling the number of sharks you only make this program available to people earning over X amount in rake each month. You can adjust the X amount every few months to keep the number of sharks at a steady number.

That’s just one idea off the top of my head. I’m people could come up with lots of variations on that basic concept.

Going to the bottom level you have all of the infrequent players you want to keep playing. A point based system is so meaningless to them as to be a joke. How is a guy who comes on and donks off $25 a month supposed to ever earn enough points to actually do anything with them? The program is completely pointless to a player like that.

What makes it worse is some poker rooms have tens of thousands or hundreds of thousands, or maybe even millions of players who have small point balances. Well, because there is a cash value associated to those point balances that shows up as a liability on the poker room’s balance sheet. In theory those players could come back, earn enough points to use them in the store, and the poker room would have to dole out the money. So, accountants keep those points as a potential liability on the spreadsheet which impacts the finances of the company.

Why are they even being offered a loyalty program at all? Why not just come up with some sort of bonus system that’s easy for them to obtain? They’re thrilled with $1 or $2 here and there. Give the guy a little meter in the lobby so he can see his progress and he might spend $50 in deposits chasing his $2 reward. Don’t confuse him with raked hands vs. non-raked hands and all of that. Just put a little goal meter in his lobby and let him track his progress.

In the middle you can come up with all sorts of creative retention based programs if you’re willing to get past rake generated and net winner/loser sort of blinders. For instance, what if you can identify a segment of your customer base that is willing to complete tasks in exchange for points or freerolls? Why not take advantage of it? Make them “Like” you on Facebook for this or follow you on Twitter for that. You could even partner with advertisers and make a profit on offering freerolls. Other player segments might react to earning privileges like access to a quality training site. Others might like negative re-enforcement like getting additional points for taking a bad beat. Whatever. It’s something you can brainstorm on for months.

The point is that then you can base the value of what you’re giving back on how valuable they are to you. For instance, if you find that you mainly get net depositors taking advantage of one type of program you can actually afford to spend more on it to attract more people of a similar type. If you get mostly hard-core grinders taking advantage of another type of program then you can adjust the points needed or how they earn points more targeted to what you expect your ROI to be.

See, in the online poker industry you have to collect massive amounts of data. Every bet, every click, every login, every action on the site is recorded in a logfile somewhere. And you can associate all of that data with a real human being. You have his address, his bank account information, etc. If you gave this kind of data to a econometrician (the kinds of guys who figure out stuff like putting diapers in this isle and beer in this isle increases profits by 2%) they would have a field day. Yet very few online poker rooms even have basic data analysts.

I even wrote a post about this way back in 2005 called Designing the Perfect Online Poker Room. In that article (remember, this is back in 2005) I wrote:

Of course getting and retaining customers is important but when the poker boom starts to level out a bit the poker rooms are going to need to increase the profitability per customer. One way to do that is to learn a little something about data mining. Based on the offers that get flashed up to me at places like Party and PokerStars, they’ve got a loooooooong way to go here. I mean, just offering the saved search feature I mentioned above would provide a wealth of marketing information. Instead of telling some guy who plays $25 NL Hold ‘Em tournaments about a $300 tourney is a waste of both the poker site’s and the player’s time. Data mining can point you to what players already want and also create profiles that can be used to segment customers. Players who spend X amount of months at the micro-limits are unlikely to graduate into better players who generate any sort of substantial rake. Once you know that, you can tailor marketing campaigns around it.

It’s chilling that I called that so perfectly. 🙂

Seriously though, in six years and billions of dollars in revenue later I’m not even aware of a single poker room that does this kind of sophisticated analysis. If you want customer loyalty how about giving them something to be loyal to? You can’t raise the player switching costs (which often results in loyalty) if you don’t even know who your customers are or what they want. You can’t tie them into your brand by spamming limit cash game players with your Sunday big money buy-in tournaments.

And the saddest part of the entire state of loyalty programs is that the poker rooms have all the data they need to construct something better. They’re just too lazy to do it. That’s why there’s no loyalty in this business.