Building an Analytics Strategy by LinkedIn Product Manager
Building an analytics strategy is crucial for every product manager. So, how do you plan, build and implement an effective product analytics strategy that quantifies and drives product success and iteration? In a recent Q&A session, Ali Khodaei of LinkedIn answered this, the audience’s questions, and gave insight to how LinkedIn operates.
Group Manager at LinkedIn
Ali Khodaei is a Group Product Manager at LinkedIn. He is also the Product Lead on feed relevance personalization and ranking of the items used as well as machine learning. He has over 6 years of product experience with companies such as LinkedIn, Microsoft and Yahoo! He has worked directly with hundreds of product managers to build sophisticated analytics strategies, and holds a PhD & Master’s degree in Computer Science.
Analytics Strategy and Building It
Ali Khodaei talked about data analytics no longer being a luxury, but a competitive and integral need for Product Managers. Product Managers need to understand not only who their users are but how they behave and know the metrics. He also discussed the importance of understanding the product ecosystem, how tools interact, and how you can position analytics from a product perspective across your company.
- Do you consider a scroll an engagement (positive signal)?
- How do you evaluate the performance of a product that is very backendy?
- What type of research do you do to understand the the user quality/quantity?
- Do you prefer to hear more about the actual problem your customers have or the features that they want to have?
- How does LinkedIn prioritize the content they show to the users?
- How does LinkedIn choose the people that they user should connect with or contact?
- Can LinkedIn use the data they collect from outside LinkedIn?
- How does LinkedIn decide what is relevant and how does the algorithm work for showing the relevant things?
- Where do you see a shortage in work force at LinkedIn in 5 years?
- What do you do to get data from your team (engineers, designers, data scientists, etc.?
- There are different models that product managers work with. Which one do you prefer?