# Calling all math weenies!

Here’s one for ya…

Person owns website.
Website is a membership/subscription style site.

What would the formula be for figuring out the retention rate? By this I mean, when the website has members that have been paying members for anywhere from 2 months to 2 years, and there are new members daily, plus new cancellations daily, how can one figure out, relatively accurately, what the AVERAGE membership duration is,using a formula rahter than actually tgoing through years of records and trying to figure it out?

Let’s assume, for the purposes of this formula, that the total membership has remained roughly constant for 12 months or more.

Any ideas?

This is the first thing that came to my mind, but I’m not sure how well it will apply to your situation.

If you know how many members you have at any given time (and assuming that’s constant), and how many members you’ve had total, you could figure it out by

(members at any given time) * (how long website has been running)/(total members)

So, for example, if you have 350 members at any given time, the website has been running for 2 years, and you’ve had a total of 2000 members, to find the average membership length:

350 * 2 = 700 man-years of membership.

700/2000 = .35 years average membership length.

Does that help in your situation?

Well, assuming each member has their own account, simply find out how long each one stayed by looking through the records, and take the average.

Or did I miss something here?

Seems like an ideal job for a computer.

I don’t know if this question yields any useful information, at least not in any business sense. I think you need some lateral thinking.
If I had to analyze this, I’d first take a sample of members that cancel, and see how long they’d been members before cancellation. You’d toss the data from people who drop immediately. Then you’d have an average subscription length from cancellers. And then look at the data and see if you can do something to keep them subscribing. Like do a min/max analysis of subscription rates. See if they would stay subscribed longer if rates were lower.
Then maybe take a random sample of longterm subscribers, do a little analysis of your existing customer base. Hope your statistics knowledge is good. Again, min/max analysis might yield data on how much you could RAISE subscription rates before you start losing longterm paying customers.
You need some spreadsheet with some statistics behind it. I used to do this stuff ages ago in Excel, but its been 5 years and the program has changed so much, I haven’t learned how to do this with the current versions. So you’re on your own with the statistics methods. Good luck.

but that’s a number I do not have (total number ever) because it would require extracting the current from the former, and a big chunk of the records are not available.

Someone gave me a formulat once that looked like:

1 + Repeats devided by Joins = Retention Rate

But I think I missed something when I wrote that down, because it doesn’t make a lot of sense to me, and I don’t understand the “1+” - if I have 100 repeating members this month, what difference would it make to add 1?

But maybe it will help some math weenie figure it out.

In case someone who could answer it didn’t see it.

This board is too damn busy!

Your gratuitous use of the phrase “math weenie” might be the reason that this thread has not been answered to your satisfaction.

Personally, I think Cabbage nailed it. If you don’t have enough info to use Cabbage’s method, you don’t have enough info period.