Interview With A Facebook Professor

Dave McClure Headshot & Class LogoLast night I got to speak with Dave McClure who will be teaching the “Create Engaging Web Applications Using Metrics and Learning on Facebook” class with Professor BJ Fogg at Stanford University starting Thursday. Below are a few of the questions that I had the opportunity to ask Dave.

What is the new class about?
Really about building engaging web applications and using user engagement metrics to figure out what makes the right set of features and right set of marketing and promotional channels to use. Using Facebook as a lab environment. We are going to use a 5-step model for user engagement:

  1. Acquisition – where are users coming from?
  2. Activation
  3. Retension – email and other things to get them to return
  4. Referral
  5. Revenue/Monetization

Are you going to be focusing primarily on Google analytics or will it also involve specific user demographics?
I think so. Right now one of the biggest issues for us to explore as the class, and really kind of interesting when involving both Google and Facebook in figuring this out is how do we get those two schools to work together and also what other tools we might use to figure out the metrics that we are trying to collect. Charlie Sheever was the developer at Facebook who implemented the fb:google-analytics code and I’ve been trying to connect with him and folks at Google analytics to see how they are progressing with that and trying to develop it further. In addition to that we are working with a couple other companies; probably Social Media and Adonomics to see what they are doing.

Those are quantitative solutions. There are also ways to collect qualitative data that are based on user surveys and sample data; maybe even doing just user monitoring and testing. It is not exclusively about quantitative data, although that is one of the more significant goals, but really more about building that model and thinking about your product development and product marketing process using that framework.

Facebook is going to have to launch some sort of analytics software if they are going to launch an ad network.
They probably are looking to roll out more in the metrics area. We are looking forward to working with them to get a better sense of where they are headed on that. It is possible that we will work with Facebook, Google or a third party company that is developing and building off of whatever Facebook offers, that we can put that stuff together. Definitely we don’t have all the answers right now. The platform itself is still maturing, developing, it’s a moving target. Even outside of Facebook, I think a lot of the concepts that I’m presenting, are ones that people are still experimenting with.

I know a lot of people who are doing general website development that don’t really have these models in place very well yet. I do think it’s really important for startups to have a simple set of metrics that they watch and actually do a lot of A-B testing around those metrics when they are building features and trying to measure deeper down the conversion funnel before marketing channels. Understand what’s working and what’s not working at bringing visitors to the site or getting users to install the app and use it on a regular basis.

You said Facebook will be involved. How involved?
I’m still talking with those folks. I’m hopeful they will speak at one or two classes. We also have some folks from Google who are going to be speaking at one or two classes. Part of the structure of the course is really that Prof. B.J. Fogg and myself will be presenting on a topic. The class is a 3 hour class on Thursdays that is more presentation then team oriented and then a lab on Tuesdays that is more self-paced development. In terms of Thursday classes we will try to bring in outside speakers in addition to our own talks and someone who is actually doing project management, doing project marketing, building a facebook app or potentially Facebook bloggers and VCs that are interacting with startups.

The goal is to give them a set of real world experiences with people who are doing those functions in startup companies in the valley. Internally with the students in the class we are trying to structure them in 3 person teams: either 2 developers and one business marketing person or one developer and one product manager/product marketing person. This is so that they really get the full experience of building a startup team and working together to put together their app and do the marketing.

What do you consider a successful application to be?
Our goals are to have each of the teams build two different types of apps during the course of the semester:

  1. One will be measuring based on acquisition metrics. The goal will be a broad use case app, applicable to a lot of people, lite-weight like a Zombies or profile badge type of application.
  2. The other type of app is going to be more depth of engagement metric. Very narrow use case probably around education topics but the goal is to see more depth of user engagement: repeat visits and retention time being the key metrics there.

We’re trying to contrast those two different styles of app development and in fact will probably be using the first set of apps to market the second ones. Also, having the class cross-market each other apps as well. That’s one of the benefits of working in an environment where we have 10 or 20 teams working together. They can use each of their apps to cross-market to each other. At the end of the semester we might be able to build an application network within Stanford. One benefit is that some of the students in the class have already built Facebook apps.

Mashable said the concept of grading based on number of users seems somewhat flawed. Do you see any issues with this method of grading?
I think they misinterpreted the grading. Perhaps a very small percentage of the grade will be based on active users but I think we will be looking for a good split between breadth of use and depth of use. In one sense we might look at active users and number of referrals as a measure of success but in others we might look at time spent or pageviews or repeat visit behavior as success.

The apps themselves are not the only measure of success. We will be looking at class participation, what they are doing or how they are interacting with their teams. There will also be an expo at the end of the year where we will bring in a panel of judges. Hopefully we get folks like Arrington and other VCs. We might actually see some applications that come out of this and get funded.

When does the class start?
Thursday. We are hoping to put together a set of course materials to place on the web at the end of the semester. We will also try to videocast some of the sessions.