“Getting Products Out From Under The MIDDLE Of The Bell Curve And Exceeding Expectations” with Preston Smalley, Executive Director, Product Management, Comcast
By Dan Galatin
September 2014 Event
Preston opened by discussing specific techniques and practices for producing excellent products rather than mediocre ones. He began by describing his experience helping to create the first app for
the Apple Store at eBay in 2008, despite the fact that the direction from executives at the time was to focus on the desktop rather than mobile. The executives “reluctantly” went along after the project was a fait accompli, and it was a great success. The moral of the story is to do the right thing for the company and the customer, to ask for forgiveness rather than permission. This project was also an example of how less is more: a simple interface can be more effective than a complex set of functionality that has built up over the years.
When products aren’t exceeding expectations, there are several common symptoms that are associated with this condition. The team often concentrates on “pet” features requested by executives. There’s lack of agreement on what to build. The team lacks analytics, or else they track everything and don’t align on a few core metrics to measure success. Finally, there isn’t enough time to sketch or prototype due to the pressure of getting version 1 out the door.
Product managers need to bring good ideas to their teams and help them unpack those ideas. Help teams understand why their building features, rather than micromanaging them. Challenge executives’ pet projects by asking what customer problems they solve, and what key metrics would change. Offer to work with the team to look at other solutions. Ask the executive how the priority of the project compares with other top priorities, and ask the executive to make tradeoffs.
Great leaders don’t tell people what to do or how they should do it. They tell them why they should do what they do. The “why” helps align UX designers properly behind the purpose of the customer problem they’re solving, and helps motivate engineers to care about the project. It helps PMs stay focused on what’s most important for the project. Preston suggested using Ash Maurya’s Lean Canvas as a technique for clearly and succinctly defining the problem you’re solving and your solution.
He described his experience working on the Plaxo Personal Assistant, a product that ended up not being successful after taking much longer to develop than anticipated. There were a number of “missed opportunities” in working on this project: not determining what customers would be willing to pay for it, not understanding what other “good enough” free solutions were available, and not focusing on an easier MVP (minimal viable product). This project was a key turning point that convinced Preston of the importance of answering these questions early.
As a general practice, one should work on incubating small product ideas as well as bigger ideas and not “just swing for homeruns.” Think of a venture capital-like model when planning your resource allocation, in which the investments are your engineering and design resources.
Preston discussed approaches to metrics for measuring product success. PMs need to define a few key metrics, know how they are doing, and share the metrics with the team and executives. Don’t believe “vanity” metrics even if you need to use them to evangelize the project. A recommended approach is to use Dave McClure’s “Pirate Metrics”:
- How do users find you? (Acquisition)
- Do users have a great first experience? (Activation)
- Do users come back? (Retention)
- How do you make money? (Revenue)
- Do users tell others? (Referral)
For a product early in its release, figuring out activation and retention is most critical. Then you can concentrate on the other areas.
Finally, Preston contended that leveraging sketching and prototyping is a core competency for product managers. There are a range of options that vary in fidelity, from sketched storyboards and wireframes, up through interactive prototypes and MVP tests. One technique is to gather user feedback on both a high-fidelity design prototype and a lower-fidelity technical prototype and converge the two over time to produce a test MVP. As you move through different planning phases, from the problem interview to solution interview to MVP test to MVP launch, you engage with larger and larger numbers of customers to get feedback. In each phase, to quote Marty Cagan, “Testing your ideas with real users is probably the single most important activity in your job as product manager.”
Preston closed advising the audience that getting better at solving the customer’s problem eventually makes you a better product manager.
Dan Galatin has over 20 years of combined experience in product management and software engineering. He is Co-Director of Communications for the SVPMA and can be contacted at firstname.lastname@example.org.