In this week’s #AskMeAnything session, Product School hosted, Shiva Arunachalam, Senior Product Manager at Uber. Shiva discussed his go-to tools at work and useful tips for starting PMs.
Meet Shiva Arunachalam
Shiva is an Analytics and Mobile focused Product Evangelist with over 10 years experience. Currently, Shiva is a Senior Product Manager at Uber, managing the Automation and Personalization stack for the Growth teams at Uber that includes Uber’s IFTTT engine, Customer persona store, Real-time trigger platform, and Experimentation.
Go-to Tools for Product Managers
What methods do you use to keep the product vision and progress sufficiently visible to all stakeholders?
- Monthly newsletters
- Vision / Strategy deck, one pagers constantly updated
- ELT / Senior leadership updates quarterly
When testing a new feature or product in a smaller market what percentage of adoption do you look for to determine it’s ready for wide scale roll out?
My experience is to do this with a controlled A/B test that shows positive signal in at least one of the metrics identified for success. Not all the metrics will be positive, but leading indicators matter more since the actual metrics may take time.
How does your team settle on key metrics for your product? What is your process for selecting one North Star Metric?
It is important to tie back your efforts to what business objective you’re trying to influence. For example, engagement, vs monetization. Once you have alignment on that work backwards to see how your product or feature can affect that and how users use that feature. When you know that better, you can arrive at a North Star metric!
How does Uber try to avoid being “feature factory”? How often do you keep checking your product vision and goals?
Test test test! No idea is a bad one, no feature is a bad feature. But you take into consideration the opportunity size, the impact it can create and the cost of building it to prioritize it. But once built, you test and only keep the features that work in the market. Kill the rest! When you define the feature it needs to align with the vision and goals all the time. The opportunity size directly impacts the business objectives.
What SDLC methodology does Uber work on?
Uber is pretty large and the teams generally have autonomy to set up the processes that work for their teams. My team uses Agile processes, and because the PMs are involved pretty intimately in the feature development they are part of the process. At the end of the day the PM owns the “What” and the “Why” of the problem. The Engineers own the “How”.
How does your team decide upon the 3rd party APIs to subscribe to? How does that fit into the overall product roadmap?
The best working model to build products I have seen is the hybrid build and buy model. Teams will never have the resources to build everything end to end. And there will always be 3P solutions that solve the problem. The way we look at it is to keep in-house whatever is a strategic priority and IP for the company. Everything else can be a 3P solution provided the cost/benefit works out.
How do you/Uber perceive product designers that have a background primarily in another industry (i.e. design, architecture, etc)?
Diversity of thought is celebrated. We have PMs who come from many different industries. What is more important is the future, how will this person add value to the company going forward!
How do you manage conflicting interests from the customers? What if the business/customer priority changes?
That is in essence why a PM exists. To keep a pulse check on how a customer’s business is changing and whether the product being built will be relevant. A great way to manage this is to build MVPs quickly, and then go deep into the feature set. The goal should always be to add value to the customers, not to build a huge feature set!
You might also be interested in: How Product Managers Assess their Products and Users
When do you know the investment in Marketing Optimisation tools is worth it? What are external analytics tools missing? Do you think these could become SaaS products that could be sold?
The one key reason why companies end up developing these tools internally is because of the need for transparency. Most 3P products are blackbox solutions and we end up relying on an algorithm with a lot of unknowns. ROI on optimization products is rather straightforward, it ties back to business impact. Are we getting a better result than without it. I would love to be in a world where SaaS companies build more transparent products.
Tips and Recommendations for New PMs
What would be the first things you would do in a company that has never had a product person before?
- Educating leadership on the value of a PM. This is the toughest part.
- Actually rolling up your sleeves and taking on more than a PM would in larger companies. Make it a fun learning experience!
What analytics and growth must-have toolkit do you recommend for aspiring product managers?
The key here is to be close enough to the event definitions to understand what the user is doing on site or app, but beyond that it’s a matter of analytics starting with:
- Engagement metrics: Sessions, Users, Actions
- Retention: Cohort views and lifecycle analysis
Most analytics tools offer these out of the box like Google Analytics, MixPanel etc. But more importantly, familiarity with SQL, and diving into the data is probably more satisfying!
Can you explain the Uber APM program?
It’s a great program for folks interested in becoming a product manager. You have a rotation program across different teams so you get to see many sides to the business. Sets you up well for the future.
Any recommendations for new PMs when looking at Data?
It is important to look at the data as it relates to the business and validate whether that is really happening. Oftentimes the data may say something but business impact may be different. Being critical of numbers is a healthy skill to have!
What would you say would be the top resources you have come across to help think like a PM?
I would say for B2C products, Hooked is a great book. A lot of product building is understanding strategy, so Playing to Win, Good strategy Bad strategy are all great books!
You might also be interested in: The Product Book: How to Become a Great Product Manager
What’s your suggestion for a transition from an engineer to a PM if moving within the organization is not an option?
You have the tough part out of the way, you know how products are built. Now move on to why do you need these products and what is the product itself. Once you get a handle on answering those questions more confidently, you’re set!
Don’t miss out on our next #AskMeAnything session for more!