When you gaze long into an abyss, the abyss also gazes into you. It’s been some time since my poll regarding the first months of new player experience in Eve. The spike of hits appears to have died down, the numbers have mostly stopped fluctuating. I shall leave the poll open just in case, but it’s time to review the results.
Out of 845 people who saw the page, for some reason, only 328 chose to answer the poll, less than half. Whether it’s because my post is unreadable, because they came looking for something else, (what?) or because they didn’t find an option that fit them but were too lazy to tell me is unknown. Maybe they didn’t cross the gap yet? Whether that selects for any specific property of their character or player experience is also unknown and we can’t tell.
Which actually demonstrates the problem with the use of surveys in sociology as such pretty well, but I’ll leave that tirade for some other time — I already spent a whole chapter of my PhD thesis beating it to death by dismantling my own results acquired one chapter prior, and wouldn’t want to go back to it.
Before you see the final results as "the player base is helpful towards new people"1 and pat yourself on the back, I need to remind you that what you’re reading is the responses of people who are already selected. We have several biases of this nature in here, but this one is the biggest — the poll results can only cover people who did get over the gap. We really have no clue how helpful or unhelpful towards those who did not get over the gap everyone has been, but observation says that ‘unhelpful’ is probably closer to the truth. It’s well outside my power to pose a question to them, because they probably don’t have any reason to read my blog.2
I can, however, estimate how many people did not cross the gap, if only by calculating how many people did not proceed past the trial. It is known that between 30k and 50k trial accounts are active at any given time. Let’s take the lower bound of 30k and assume that every trial account is active for the longest possible time, 21 days. It is also known that Eve grew from 302k to 352.5k paid subscriptions between October 2009 and March 2012, making it a growth of 50.5k in 2 years and 5 months.3 We have no clue how many of those have been alts, (observation suggests a bloody lot) but let’s assume for now that all of them have actually been new players.
With the above assumptions, during this time period, ≈1.2 million people tried Eve (but probably quite a bit more) and did not purchase a subscription, as opposed to the 50k people who did. (!) That’s a 4% new player retention rate at the very best.4 Since the growth number does not account for unsubscribed users or alts, and many trial accounts are not the full 21 days, the actual retention rate can be higher or lower, but it’s very definitely within single digits, and more likely to be below the 4% I got rather than above. For every one person that remains, at least 25 people leave empty-handed.
I don’t know when they actually leave. CCP does, down to the exact hour, as Hans Jagerblitzen tells me. CCP doesn’t actually tell anyone. From what I can see, it looks like they apply the Microsoft approach to fixing bugs to the problem: Find a problem experienced by 80% of your audience, no matter how severe or minor, and fix that first, and you cut the number of bug reports by 80%. Find the moment when 80% of your new users leave and fix it, and you’ll have 80% more new users. Which makes good practical sense, if you’re talking about software bugs, but doesn’t actually work when applied to this particular problem.
Well over 60% of those 25 people most likely drop off in the first hour, put off by the gameplay they don’t like, the very idea that will never, ever attract them no matter how much you polish it, just because it isn’t what they were looking for, determining which doesn’t take long. Chasing after them is futile — just like the people who saw my poll and chose not to answer it, they will be lost, because the game doesn’t fit their niche, and you can’t do anything about that. Shouldn’t even, because you might scare off those that would otherwise remain. What should concern you most is people who played until the end of the trial but never subscribed, or dropped off after having subscribed for less than three months — the people cut off by the gap.5 The remaining 10. We can’t really find them to ask, they’re already gone. But we can find out what did get people across the gap and see if anything can be done to make that happen more often.
With that in mind, let’s look closer at the numbers:
The sociable: 43.29%
I found an established group that I had no prior acquaintance or unknown prior social connection with. They took me in before I reached 5 million SP. 35.06% (115 votes)
In retrospect, wording this option like that was a mistake on my part.6 There are corps that impose lower SP limits, like 1 million (30 playing days) or 3 million (minimum practical competence level with tight specialisation) or employ other mechanisms, like special sister training corps which offer basic training for candidates, but only move the candidates up from there upon reaching a certain level of character development. All of these fall into this option, and now we don’t know their relative weight, which would be interesting to know.
Nevertheless, that this is the most common option means that corp recruitment actually does sort of work, and covers about 30% of all people who cross the gap. That’s quite a lot. In the ideal case, however, it should cover ≈80%, as this would allow for the optimum player replacement levels and maintain stronger in-corp cultures. Why it doesn’t is something to look into.
I struck out on my own with a group of players just as new as I was or close, whether before reaching 5 million SP or after. 8.23% (27 votes)
I’m actually part of that group myself — as the new expansion rolled in, I saw industrial opportunities and enticed people from the new player corp to join me to exploit them. Since for me, leading a corp is quite abnormal, I quickly burned out and missed a year soon afterwards. The corp still exists for some reason, but none of my corpmates from that period still log. What I’d really like to ask the people of this group is:
- Does this group still exist? If it doesn’t, for how long did your group survive and how did it die? If it does, and you aren’t a member anymore, why?
- Was this a highsec, lowsec or nullsec group? Industry, PVE or PVP?
- When exactly this happened?
It is my hunch that the suitability of the environment for such new player groups has drastically changed over the years, and not for the better.
The well-connected: 22.56%
I was invited to Eve by a more experienced friend/acquaintance, and they kept my interest by involving me in their game from the start. 14.33% (47 votes)
Not at all surprising, but considerably higher than what it should be in the ideal case. (I’d say 10% would be the right number for a niche game like Eve.) Probably the happiest option, to be honest, and certainly the most independent from all other factors, but it relies on having a tight personal connection. I tried several times, but never managed to pull it off, I do wonder how many of those invited personally in this fashion drop off.
I already had an established group I expected to become a member of, because I knew them before signing up, and joining them did not involve SP membership requirements. (TEST, Goonswarm, etc.) 8.23% (27 votes)
Hi Reddit. Almost all of the people who selected this option are obviously TEST members. I honestly expected them to be more common, because ≈95% of the hits to this poll were from Reddit. But in retrospect, relatively few TEST or Fweddit members create characters explicitly because they know they will be TEST or Fweddit members upon creation, and the same is probably true of Goonswarm. Quite a bit more create characters first, then go looking for other people who play Eve in their preferred offworld community, then join, and thus end up in the "sociable" category, when they’re probably better placed here. (Which means the corp recruitment actually works much worse than it would appear to work from these poll results.) I already spoke a lot about how an offworld trust base is a tremendous advantage in Eve, no matter how weak it is, but with this poll’s low and narrow exposure, it doesn’t really provide any kind of strong evidence to confirm or deny my words. Oh well.
The loosely connected: 13.1%
I kept playing on my own with social influence of a new player corporation or a help channel well past the 5 million SP watermark. 2.74% (9 votes)
I mostly played on my own, but the social influence of my new player corporation or a help channel was the deciding factor to keep playing until I reached 5 million SP. 5.49% (18 votes)
I expected those two to be slightly higher, maybe up to 10% total. I’m not sure how much would it be in the ideal case, the weakly social options should pick up the people who didn’t fit into the strongly social options. Back when I was new, University of Caille was a stable carebear haven, where you could always find conversation, advice, and sometimes, actual help — certainly, you could find other players just as new as you are to cooperate with. Then the NPC corp tax was introduced, and the most hardcore carebears which made it a thriving community quickly vanished into shell corps of their own. New player corp channels were pretty much nothing like this since, and probably wouldn’t motivate anyone anymore. Right now, I know only one help channel which forms an environment of this kind — MOZG. Not surprisingly, it’s like that because it’s Russian, and Russians have always felt linguistic and cultural pressure in Eve.7
I joined a large loose and open group with no membership requirements, which allowed me to get noticed by a tighter established group that eventually took me in. (RvB, Faction Warfare, etc.) 4.88% (16 votes)
This should probably be slightly higher, too. I definitely see very new people in Militia chat, and they do appear to eventually survive more often than not. It might be, though, that the reason this option is so uncommon is that low standing requirements to join a militia are a comparatively recent innovation, just like RvB has not always been widely known about.
The stubborn loners: 21.03%
I used regular currency to offset the SP/resources starting disadvantage through the purchase of PLEX/ISK/character(s), which allowed me to join an established group earlier. 1.83% (6 votes)
I wouldn’t call this strong evidence, but it shows that people who actually solve the social problems of Eve by throwing money at them are probably not common, despite popular belief. Which, I think, is a good thing. :) Needless to say, this shouldn’t happen at all — if playing the game requires to sink so much money into it to get in, it’s not accessible enough. Comments on Reddit seem to indicate that it can be a problem.
I kept playing on my own without social influence well past the 5 million SP watermark. 13.41% (44 votes)
I played on my own without social influence up until joining an established group upon reaching 5 million SP. 5.79% (19 votes)
That’s mostly the People Who Like Spaceships, who preservere alone despite the hostile environment because they like the setting. They’re the people who were in the most danger of having dropped off while in the gap. In the ideal circumstances, almost all of them should have been members of some other category instead. That they aren’t is a credit to them, but to Eve itself it’s a crying shame.
There’s something I wish to ask these people too:
- So what kept you from joining a corp earlier? Did you not find a corp you like? Did you not find a corp that liked you?
- What exactly kept your interest? Was it the promise that you will eventually find someone to play with, the natural properties of Eve that you feel strongly about, (spaceships! explosions! nebulae! complexity! etc…) or something else?
- If you still play alone, for how long did you play?
Yes, commenters on Reddit, I’m talking about you.↩
But if you are a non-Eve player reading this, who tried Eve and failed, please come forward and tell us of your experience, it’s important.↩
For comparison, Second Life, which is so similar to Eve Online, had 16 million accounts the last I checked, (no concept of a trial account there) which they frequently mentioned, and of which only about 100k were the core population and 300k were the population seen as ‘active’ — quite close and within the margin of error for this eyeball calculation.↩
And as everyone knows, we must not allow a mine shaft gap!↩
Whoever does that email survey for CCP, if you’re reading this, take note, you do that more often than me. That’s why you do a pilot survey and adjust your questions — treat this as one.↩