I upvoted just because I'd like to know myself and to hear your guys' thoughts =). We're always open to discussion with our potential customers.

Thanks for the upvote :)

I'd really like to use Mixpanel, however I fully understand your decision on not allowing historical data. Also, I appreciate how responsive you've been so far.

http://kissmetrics.com/ is the most similar product.

R is a programming language with huge number of data analytics libraries. It's not really an alternative any more than, say, Python is an alternative.

I've used R before, exactly for graphing some data for this app, so I can say - no, it's not like Python or any other general use language. You can't really do a scatter plot from a csv file in Python in one line.

For example, have a look here: http://www.statmethods.net/graphs/scatterplot.html and observe how little code you must write to get those graphs.

We switched to KissMetric's beta product and have been pretty happy with it. Doesn't support importing data though I don't think.

I'm pretty sure you can with KISSMetrics. They have a PHP/Ruby API where you can set the dates for stuff. Just run all your old event data through the api setting the dates when they actually happened.

You can, but we'd really like to talk with you to understand more about what you're trying to do (if it's a valid use case for you, it probably is for others, and we want to make sure we get it).

Email calvarez at kissmetrics.com and I can help you get started.

I don't believe kiss lets you import data either but maybe there's a way.

I don't really understand your problem fully, but if Mixpanel looks like a good fit I would recommend using them for your new data, and creating a separate system for viewing your historical data.

That reduces the requirements for the code you have to write (it only has to be able to deal with historical data, you don't need to figure out how to deal with updates), while giving you maximal features in the long run.

My problem is rather simple actually. We currently track events in a local database very similarly to the events we would send to the Mixpanel. These events have been analyzed only superficially until now. If I wrote the code to analyze and plot the data, it would work on new data also and would make Mixpanel somewhat redundant (dealing with updates is a non-issue, it's the logic that's difficult, not dealing with more data).

However, I don't want to reinvent the wheel, or to invest a lot of time in something that other apps like Mixpanel, NuConomy already do very well. I suspect Mixpanel uses the timestamps of the events for billing and if so they might require some non-trivial changes to their business logic to allow historical data. I hope they will, someday. Until then I might settle on sending them only new events and using R on the historical data.

I'm not sure what trends you are looking at, but most consumer startups usually find it unnecessary to look at historical data points past X months. Things (features, users, etc.) change so much that averaging in those data points so far back would misguide you at the present. I would go with a solution that is almost perfect for you if you started today (for example, Mixpanel might be it) -- gives you the best odds for your future analysis. During the couple weeks after switchover, I would just export the Mixpanel data to a local spreadsheet (and run in R as you mentioned) with your other data. After sometime, you won't need it.

I'm not aware of any service that lets you import historical data.

For a start, each of these services collects a slightly different dataset measured in a slightly different way-- and so if you wanted historical data to analyize trends, you might be comparing apples to oranges with the data from the old system vs the new one. I'm also not sure there is a standard format to import/export this kind of data.

What system is the historical data sitting in right now?

The data is in a mysql database (persisted from an ActiveRecord event model). We track things like what was done, by whom and on what object (i.e. user_id, target_type, target_id, action and other stuff).

What we need is a system that goes through all these events and lets me plot a graph with logins/day for users with a free accounts vs users with a premium account (for example). Mixpanel does exactly this. There is no need for any kind of standardization on the data. Just go through all these events with properties, and aggregate them to plot interesting graphs or to gather other actionable data. The only real requirement is some id to aggregate these data points on a per user / per account basis, and a timestamp to aggregate them chronologically.

I'v been using http://wingify.com/ and have been quite happy with it. It's by a fellow HN member. Not sure if it fits your exact needs, though.

Thanks for starting the thread, I would be interested in knowing what you eventually decided to go with. I too had been running our own analytics for a while and started looking at Mixpanel to see if we could possibly migrate to mixpanel in order to make more sense of the data we've collected and collecting.

There is Vanity for Rails: http://vanity.labnotes.org/ It does not support funnel analysis like Mixpanel tho.

I've been looking at it yesterday. I've already integrated ABingo just last week. Any thoughts on pro/cons between the two ?

Metricly seems to allow importing historical data. However, I suppose being in a private beta is a drawback.

http://chartbeat.com/ is pretty impressive.

mtarnovan, I can try to help. Drop me an email at hrishi@prettygraph.com.