Userbase Analysis

[Tad Carlucci]

We're almost to the 2-year anniversary of the first account on this web site: December 9, 2010. That prompted me to take a look at how it's doing. Well, that's my story, and I'm sticking to it!

Accounts on this web site are a good measure of how many people are actually playing. Their performance, over time, can show whether the game is vital or failing.

Now, without writing a program, which I would consider abusive, it's not possible to collect data on every account. So I settled for a random 5% sample. Since, at the moment I checked, there were 7,780 accounts, that meant a by-hand survey of 389: a number small enough to do in an afternoon.

This might seem like a small sample, but it's enough to gain over 80% confidence that the sample realistically represents the entire population. There will be some error, nonetheless. For example, I think a fully random sample actually skews the numbers a bit. Why? Because early-adopters are more likely to hang on for a long time than those coming along later. So, while I'm fairly confident the survey is accurate, I think it slightly under-represents the effect of those early-adopters. Similarly, it unfairly represents the most-recent additions. Fixing such errors is impractical, the effects are probably small, and probably balance each other out.

The chart below shows the number of web accounts, over time.



Two things immediately jumped out at me: there are only two phases, and both are very (and I mean VERY) straight lines (as this sort of thing goes).

As I recall, KittyCatS opened some time in February, 2011. From opening, a fairly constant 4.25 new accounts were created each day. Some time around mid-June a major shift occurred. Virtually overnight the rate changed to 13 new accounts being created each day.

What is remarkable is that these rates are SO constant. I wasn't surprised to see it growing. It can't go down. And every day you see a new-user greeting at the bottom of the main Forum page. But .. it's SO constant!

But, all those new accounts are not actually players. A number of them will be people who come, take a quick look, decide it's not for them, and never return. So, took seven (7) days as the break-point. If someone was only around for less than a week, they could not possibly have bred a pair and gotten a box. So I declared them "just looking".

My sample indicates that fully 65% of all users exceeded that point! This is AMAZING. A new-player retention rate of 65% is phenomenal! It's about TEN TIMES what I expected. This speaks directly to the success of the KittyCatS marketing plan.

But, the real question is, having enticed someone to come look, and having managed to get 2 out of 3 of those to actually stay and play .. how many of them are still playing, today? For that, I chose a 30-day cut-off.

17% of all 7800 web accounts are actively playing today. Put another way, 26% of those who came and stayed are still playing today. Or, since I'm often asked to put a hard number out .. as of today, approximately:

7800 people have "checked out KittyCatS", of those
5100 people have "played KittyCatS", and of those
1320 people are "currently playing KittyCatS"

HAPPY BIRTHDAY, KITTYCATS RESIDENT!

[Sanura Snowpaw]

I think you were bored and just wanted some stats to look up and figure out

That said it is still very interesting stats.

[dakillakm Resident]

So, if I understand correctly, it looks like KC is pretty good at getting people to breed for a while but may still want to be working on long-term retention?

[Tad Carlucci]

Most breedables seem to be characterized by boom/bust cycles based mainly on the timing of the introduction/discovery of new trait values, frequency of production, and the effective length of their breeding lives. The boom times draw in new customers. If the bust periods are allowed to become protracted, there will develop a "is that it?" attitude and customers will begin to drift away. My point, I think I'm making it mainly on another thread, though, is that the KittyCatS products are best at pulling in new people with their Yule/New Years introductions, and weakest over the summer doldrums. If it were my company I'd plan ways, now, while there is plenty of time, to address the summer doldrums and I'd be watching the upcoming season to see what works and what doesn't and use what I learn to try to spread the good effects throughout the year.

---

The big thing on this thread that I want to stress is the KittyCatS is not simply "doing well" .. they're doing phenomenally well! Retention rates such as we're seeing here are vastly higher than what I'd expect.

What I see, is a product which is able to catch and hold customers. I would be looking for ways to expose more potential customers to the product lines. Given the monies involved, I would probably approach Linden Labs about some form of co-advertising on cable networks which appeal to the main demographics .. probably something like HGTV or Lifetime Networks .. rather than working inside the SL environment, itself. That would be a win-win for everyone, including SL merchants having nothing to do with breedable products.

[Brandon Dawson]

Awesome data, appreciated!

[Saga Felix]

Without having any data or statistics, I had sort of figured out that KittyCatS have been growing very steadily ever since I myself started breeding. What you did there confirms my "gut-feeling" in an awesome way that I could have never done (or well, only if I was extremely bored and with great difficulty, lol) - it's great to have an analytic mind around

Thanks, Tad

-Saga

[Kayleigh McMillan]

I have a question for you Tad.
Did you take in consideration people like to use alts to sort their breeders and to have multiple hunt items for example if they are ristricted to one per person?
It seems to me most KC users have alts for this purpose, not me ofcourse *angel*, this could explain not only the amount of KC addicts, from which I think a lot are fictive, but also why they are not online much.
Last time I tried to log more than two accounts my computer almost litteraly started to smoke and produced unusual sounds this was when mesh hit the grid.
I think the last log in time could very well meet the last group hunt time therefore.
As I recall there was a hunt in July for which a group tag was needed.

[Tad Carlucci]

There is no way to determine whether a Second Life Avatar is an alt for another.

What defines an alt, anyway? Tad is an alt .. I just never openly use my primary account any more. I have a couple more I use for specific roles. Same for Charm. Between us, Charm and I have seven or eight accounts (there's probably more but some passwords have been lost over time). And all our computers know the passwords to all them. There are times when I use Charm's account, does that make her an alt when I'm using it? As far as KittyCatS are concerned, there's three accounts for the two of us. Would you count that as 1+1+1, 2+1, 1+2, or 1.5+1.5? Or are all of them alts because, even though they're using several IP addresses on our local network, Second Life and the KittyCatS web server see them all as sharing the same address?

What I counted was web accounts. The way KittyCatS is set up, there's a one-to-one correspondence to Second Life Avatars. Other than that, it's impossible to tell.

[Kayleigh McMillan]

(12-04-2012 08:07 PM)Tad Carlucci Wrote:  ...

What I counted was web accounts. The way KittyCatS is set up, there's a one-to-one correspondence to Second Life Avatars. Other than that, it's impossible to tell.

I think you are right it is hard to tell if not impossible how many unique individuals the KittyCatS community counts.

P.s. To answer your question what defines an alt to me is: one person is using one or more virtual identities to use, in this case, the KittyCatS product for what ever reason.
Wether it is to sort cats, having access to more than one hunt item or saving cats without food in the online cattery i.e. special editions one stores for later use.
Or to have some cats on food and not milk for instance, in the online cattery, if they don't fit in a current breeding program or if they are less important at some point.
Especially since the online cattery doesn't facilitates groups of cats which we can set on non food or food only if food and milk is available etc. so these extra web accounts are needed in some cases to sort this out.
Or just alts for funsters or peace ofcourse
But mainly the group hunts limited to one per avie, which result in a greater amount of group member I think, and to tweak the use of the online cattery makes it almost impossible to tell how many real unique identities play with these cats.