Not All Online Poker Traffic is Created Equal

Dominik Kofert recently emailed me to make me aware of an analysis document they had produced over at PokerStrategy.com. Dominik is someone I’ve had the pleasure of working with on a few things while I was at Party. He’s probably one of the brightest affiliates in the market today. Not because he’s better at getting customers than other affiliates (which he is) but because he understands the entire poker universe. He understands player value and how the whole poker ecosystem works so when he discusses adding value he means real value and not just MGR (monthly gross revenue).

The PokerStrategy article starts off making a central point that there are really two types of affiliates out there, contributing affiliates and cannibalistic affiliates. As the names might imply, contributing affiliates create value while cannibalistic affiliates “don’t actually generate unique or new traffic.” That was exactly the point I was making in my post The Sky is Falling for Poker Affiliates.

While the paper points directly at rakeback sites I happen to believe that many other types of cannibalistic affiliates exist. For instance, in the the same blog post I just mentioned in the last paragraph, I discussed the fact that many of these sites optimizing for keyword phrases like “Full Tilt Bonus Code” are by default, acknowledging that the player already was familiar with the Full Tilt brand. Why should Full Tilt pay the affiliate when the player was going to end up on Full Tilt anyway? PokerStrategy makes a similar argument:

It pays the first time when it invests capital into creating a strong brand that gets recommended by word of mouth or in forums (cost for creating good software, providing tables full of fish, having a good customer support, famous pros, etc.). It then pays again when it advertises via print, television or internet. Finally, it pays for the players once more when it forces potential VIP players who were attracted to them through their initial marketing efforts, to sign up via a rakeback affiliate (that has ironically done very little to bring the players to the room).

The paper then goes on to give probably one of the most easily understood explanations of why high raking players are not as valuable as they appear to be.

As most networks have already realized, “rake” and player value is more complex than simply looking at MGR numbers. Rake is the product of liquidity which has been put in motion by being played with. Liquidity is therefore the direct pre-requisite for rake, and without a continuous influx of fresh liquidity the liquidity pool would dry up and no new rake could be produced. Liquidity is generated by players who deposit more than they cash out, contrary to most high volume players, who generally cash out more than they deposit. The truth of the matter is that the bulk of high MGR players are paying their rake with the money they win off losing players. It is these losing players that are the true source of rake, while high volume players simply ensure that this liquidity is converted faster.

This is best illustrated with an example: If player A deposits $100 and plays 100 hands in which he produces $5 of rake and loses his entire deposit, he adds $100 into the room’s liquidity pool which will later be turned into rake by other players. His true value is therefore much closer to $100 than it is to $5. So even though he rakes only a fifth of player B, who deposits $100, plays 500 hands and wins $100 while producing $25 of rake and then cashes out $200 – he is essentially more valuable to the overall ecosystem.

If you’re actually serious about running a poker room figuring out what that value is is somewhat the holy grail of poker room management. You can come up with a lot of models to show you things like if you have X million in deposit in player accounts you have an expectation of making Y in rake but I don’t think anybody has really cracked the secret of getting to a number that allows you to determine how the value of $1 of rake is. As the article implies, the value of player A is closer to $100 than it is to $5 and the value of player B is closer to $25 than it is to $100. Player A, while generating little rake adds something beyond the rake to the poker ecosystem. Meanwhile player B generates more rake but his withdrawal removes a value greater than the rake he’s generated.

Where I think Dominik and the PokerStrategy crew go slightly astray in their paper is as I mentioned before they seem to focus on rakeback affiliates as the core issue. As a result when they go on to discuss solutions there is a tendency to tackle the rakeback problem rather than the bigger picture of cannibalistic affiliates which I think is broader than they define.

Rakeback in itself is not a problem. One could make an argument that PokerStars’ loyalty program is nothing more than rakeback disguised as something else. That is why I have a little problem with the solutions PokerStrategy puts forth.

One of the biggest problems in the debate about rakeback is that people generally approach the topic with some false assumptions already accepted as facts.

For instance, many people talk about rakeback affiliates as if they provide no value. PokerStategy’s article suggested that there are contributing affiliates and cannibalistic affiliates and there is an automatic assumption that all rakeback affiliates are cannibalistic. What is the threshold for the difference between contributing and cannibalistic?

Many rakeback sites have original content, strategy, etc articles on their site. Some poker training sites are rakeback sites. How are any of those different than say, some guy who slaps a bunch of banners up on a website and pays someone to write some poker room reviews? Isn’t the room review guy just as or more cannibalistic as a rakeback affiliate who regularly puts up new content? Why is there no outcry from the affiliate community about the bonus code and room review sites that far outnumber rakeback sites?

Or where do you draw the line with incentives? PokerStrategy’s free $50 bankroll is an incentive many affiliates have no access to. Or what of “exclusive freerolls”? Are they not also other ways of returning cash to players that is not part of the centralized system where all affiliates are on a level playing field.

I don’t mean to be critical of PokerStrategy there. I think their service has a wonderful retention model. It works for them and it brings in players with good value. But at the same time, it goes to show that one affiliate model does not fit all.

The problem with rakeback today is that both affiliates and many poker rooms are slow to realize that players are demanding something better. Just on principle alone they’re objecting to the fact that by clicking on some some affiliate link three years ago some guy they don’t even know is making 35% or 40% of the $5000 a month they generate in rake.

I’ve suggested this a few times in the past but it hasn’t really caught on. The way to fix the problem, both in churning players on network sites as well as rakeback in general, is to slowly offer the player a percentage of his rakeback over time. If you sign up with an affiliate, maybe the affiliate gets his 35% – 40% for X months and then each month after that the affiliate’s percentage declines and it’s funneled back to the player. This rewards player loyalty because they have to earn their rakeback via longterm loyalty and the affiliate is paid a percentage long enough that he can recoup his costs plus some profit.

At the same time affiliate managers should start weeding out the truly cannibalistic affiliates. Many of the contributing affiliates would benefit via less competition and they could be partially compensated for the loss in revenue by phasing them out of the lifetime MGR. It probably won’t be 100% compensation but it should help.

And for contributing affiliates who really do want to help with retention and keeping players playing over long periods you can work out a deal whereby the poker room sells access to that affiliate’s special promotions. For instance, let’s say I’ve been with Full Tilt for one year and after one year the affiliate no longer gets anything. If I still want to be a member on their site or participate in their freerolls, rake races, or whatever, I have to tell Full Tilt I want to subscribe to that affiliate’s promotions. In exchange for a flat fee or a percentage of my MGR I can then participate as if I was a part of their site.

The big upside for the poker room is that they can give the customer exactly what he wants. If the player signed up on PokerNews and later he decides he would benefit more from a training website like CardRunners than the somewhat basic strategy articles on PokerNews he can switch over to CardRunners. This ensures the quality of the services being rendered by the affiliates and is a strong disincentive for cannibalistic sites.

I know a lot of this isn’t what affiliates like to hear but it’s the best situation for players and at the end of the day that’s who we’re all supposed to be thinking about.

photocred to goodnight_photography