Leave a comment

What are the ingredients for a Lean mobile analytics setup? What are some of the common failures of the 'data driven' approach and how can these risks be mitigated? What analytical approaches and capabilities will help drive growth at different stages? Andy Carvell (ex. SoundCloud retention lead) tackles these questions and more in this latest essay.

  • TC

    Tim Chard

    over 2 years ago #

    I don't even market mobile apps, and I learned a ton from this.

    "Ultimately, marketing analytics is about understanding and improving the value (as a function of cost and impact) derived from specific channels and campaigns."

    So easy to get lost in making sure the data I'm reporting is accurate that I forget whether it's VALUABLE

  • AA

    Anuj Adhiya

    over 2 years ago #

    @andy_carvell Loved this post.
    What's your take on focus on the relative "thickness" of each of these layers depending on stage of the company?
    Which one(s) are needed and/or needed more when you're, say, pre-PMF, post PMF or whatever other descriptors there may be for company maturity?

    • AC

      Andy Carvell

      over 2 years ago #

      Great question, @anujadhiya :) Tbh, this probably deserves more thought and a follow-up post. However, my quick thoughts on this are:

      Pre Product-Market-Fit:

      Investigative Analytics are massively important and the thickest layer (they tell you how users are interacting with the Product and help you evaluate when there is a 'fit')

      Marketing Analytics is maybe thin, but already present: iterating toward PMF happens from both ends (Product and Market), so it's already important to be experimenting with targeting, creatives and channels.

      Operational Analytics is thin, but present in terms of crash reporting and other basic error-catching. Possibly these errors can be caught as events fired to the investigative analytics, so might not rely on a separate solution.

      Dashboards might not be needed at all, or else might exist in basic form in the investigative analytics tool. The team might be small enough that everyone is looking at the same few numbers or funnels in the investigative analytics.


      Operational Analytics grows substantially (outages cost you more when you have more users to disappoint)

      Marketing Analytics grows in line with performance marketing activity (but probably not much if product is growing via virality and 'organic' channels)

      Dashboards become necessary to keep everyone on the same page and satisfy external enquiries.

      Investigative Analytics is still important, but grows slower: in line with product changes & new features, which will need new tracking. A/B testing framework probably added at some point (which counts as investigative analytics)

      • AA

        Anuj Adhiya

        over 2 years ago #

        If these are the quick thoughts, I can't wait for the actual follow up post :)
        Thanks for the super response - incredibly helpful.

Join over 70,000 growth pros from companies like Uber, Pinterest & Twitter

Get Weekly Top Posts
High five! You’re in.