Predictive Churn Rate Modeling

by fabio on 15/08/2013

What is Churn Rate?

Churn rate is a measure of customer attrition. It is basically defined as the number of customers who discontinue a service during a specified time period divided by the average total number of customers over that same time period. In the most cases, churn rate refers to product or services based upon a contract/fee business model with recurrent purchase, hence telecommunication, insurance, banking etc.etc. Churn rate is essential in order to calculate subsequent KPIs live Customer lifetime value, ROI and marketing mix modeling, every contract-based business has to be able to calculate this at any time.

Why use modeling instead of actually tracking customer’s behavior?

Churn rate modelling can be used in those instances in which there’s no technical way in order to actually track their behaviour or it is actually possible, but a time lag between the information being sent from the users and its collection and crunching can present some delay, so that the insight acquired wouldn’t be useful anymore.

The biggest benefit of relying on a churn rate modeling is that provides a predictive framework in order to take a look to the future and do some more predtiction work for other related KPIs, from total revenue to ROI and Customer Lifetime Value.

So far churn modeling has been focused on data crunching and correlation: providing

Why predictive modeling works?

Working predictive churn modeling work when we’re able to find the right one for the right business. There is no magic involved, there’s plenty of models out there and you have to find the right one, that one that fits your specific business case or your industry. Models based upon scientific phenomenas mainly works because those patterns are already being studied in alternative contexts, hence providing a good scientific footing for any upcoming theory.

 Uranium Decadence equation for churn modeling

With this model we tried to figure out how to predict installation and uninstallation of a certain software we distribuited. The churn rate was meant to grow as installation grows but in a completely linear fashion.

The equation for the U238 decadence is a linear one:

Uranium – 238 –> Pb – 206 + Alpha + Beta

Starting off from a 6 months data set of installation and uninstallations, we run a test with the equation and the data that was available at that time:




The result was pretty impressive: the equation was returning a pretty realiable churn rate over the base line with a 5 to 8% tolerance margin, pretty accurate for being just a random model.

Now we’re able to base more predictive work on this number since we have a model that we proved that work, so all the subsequent KPI can be based on some sort of reliable ground, even inhereting the 8% tolerance margin.

There’s plenty of opportunities out there for models like this. The won’r solve your lack of tracking problem, still you can exploit them for the time being and, once proven to be effective, they can be used to have a look towards the future and run the same prediction on other relevant kpis.

No Comments

Maximize profit and improve your pricing strategy with a/b and multivariate testing

by fabio on 26/04/2013

We usually hear about a/b and multivariate testing hen referring to UX/UI improvements, user engagement and conversion optimization. These are obvious applications for these tools, allowing users to actually test changes in real time, or even before a crucial functionality roll-out.

However, for those cases in which there is no crm backing up the pricing strategy and generating some kind of pricing insight, we can actually use these tools to find out the right price range yelding the highest revenue per clic, thus helping us to maximize our ROI and online advertisement spending. I focus on this kpis because that very price, if referring to on/offline business, could not provide the same advantages in a offline environment: revenue per visitors refers strictly to the online side of the business, linking online advertisement spending to actual conversions.

In this chart we relate revenue per visitor with the price of the product adquired through the online channel. We tested different prices on a 2$ increase scale, and tracked down revenue per visitor. Considering that a lower price generates a higher conversion rate, revenue per visitor goes up, until elasticity to price seems almost irrelevant. Then prices still goes up, but the conversion rate decreases and so revenue per visitor.

Tracking the correlation in between price and revenue per visitor we can actually maximize overall profit, finding the right price (or the right price range) yelding the maximum ROI per visitor, hence giving us a pretty clear idea on the elasticity to price of our prospect.

Base price and upselling optimization
However this can turn pretty tricky if we consider a base price and and upselling price. The right price range for your base product might not be the same for the total upselling price. This case is pretty common in all those industry which strongly relies on ancillary revenues (travel and accomodation) and generic upselling (banks, insurance). In all these cases, there’s a different elasticity to price depending on product category and purchase intent, so this can affect directly your pricing strategy.

As for this chart, the optimal price range for the base product it is not the same as for the total upselling product. All this can be achieved pretty easily with a multivariate testing. At least a few hundreds of conversion might be necessary to achieve some statistical relevance, however most a/b testing tools already take care of this.

No Comments

Web perfomance optimization: how to render a web page under 1000ms

by fabio on 11/04/2013

This video show how serious Google takes the performance optimization issue with webpages. This is basically a comprehensive explanaition of the methodology on how build quicker mobile website There’s also some really valuable insight which confirms the correlation between loading times and conversion rate, pointing out web performance as a crucial factor in increasing revenues.

Besides that, it makes you understand quite well the strong drive Google employees have for constant optimization, something you don’t see in that many companies after all…

Here’s the full presentation: building faster mobile websites

No Comments

Should I invest in Bitcoin?

by fabio on 9/04/2013

I’ve been researching a bit the topic on my own and the bottom line is: NO.

In the last two months the value has increased by 10-fold and in the last year the value has increased by nearly 100-fold. Bitcoins are exhibiting tell-tale signs of a bbubble, just like the Tulip Bulb Mania in the 1600s and other countless bubbles since then.

We are currently at the point where average people re starting to take notice and think, “Maybe I should get in on this too” and start investing money into it, further perpetuating the bubble and giving it it’s last big expansion before it finally bursts. If you do want to get in on bitcoin don’t do it right now at the end of the mania, wait until it crashes and buy when it’s low again and everyone who thought they just became near-millionaires are licking their wounds from greedily holding on to their bitcoins too long.

Bitcoin Resources
BitCoin Talk
Huge Repository of Bitcoin Resources

No Comments

If only I knew this shit in college

by fabio on 29/03/2013

Probably because I am right in the process of making quite an important decision, this presentation resonates so well with me right now.

Throughout all our formal education years we grow a distorted and heavily biased vision of life and work, we build up absurd expectations and this eventually leads to equally absurd delusions. I’m referring to the idea that a multinational company might be the best place to work, while you soon discover that a small, agile and hungry for success startup might really be the best place to work, despite salary and other conditions.

This presentation flawlessy explain that, I should definetely go back and read it again every single time I find myself wondering about what might next step should be.

No Comments

validate your sitemap for errors with sitemap inspector