Break out of the local maxima

This is part 4 of the 4 part series: Growing eCommerce revenue $0 to $200MM at TripAdvisor

Lesson 4: Break out of the local maxima. The next stage of product growth that follows a period of intense test-driven optimization is stagnation.  Unfortunately, the closer you get to your local maxima, the statistical significance of your tests will become increasingly . . . insignificant. 

1. The spotlight made fade. Everyone loves a new product launch. Internally, and externally, the days and weeks leading up to a new product launch are filled with excitement, anticipation, and a lot of hard work put into plans that, by definition, are meant to have meaningfully positive impact to the business. That excitement spills over into the those first few months and quarters after the launch as the team rollouts out planned improvements, closes off any loose ends, and basically grabs the low hanging fruit of optimizations. Over time, the new thing becomes the old thing and public attention wanders. If this happens, it can feel like a big let down, and it’s important not to take this personally. This is not necessarily indicative of loss of support for the team, just more a reflection on the power of a “shiny new thing” in our collective conscious. 
   
2. Expectations are reset. Grabbing low hanging fruit is easy and fun, and your team (and bosses?) have gotten used to a period of success. But much like everything in life, external factors are much more at play than we realize. It doesn’t take too much effort and the size of gains from an individual win can be large. As the low hanging fruit  it gets a lot harder to find any sized win. It’s important to change your expectations to match this reality. If you are hoping to find your next 20% win and “only getting 1 or 2% wins all the time,” you are setting yourself to be constantly disappointed. 
   
3. New team structures. At TripAdvisor, two of the best product teams in the company sat next to each other. One of those teams was excellent at optimizing what already existed but was terrible at reinventing the wheel and would easily get caught up in paralysis analysis. The other excelled at launching large scale, technically complex products but would get bored with the perceived monotony of weekly test-driven optimization. In truth, both teams were necessary at different moments of a single product’s life cycle, which is why they sat and worked so closely together.  

If you find yourself in a situation where it’s been a long time since the team’s last “win,” it may be time to consider that current experience you’re optimizing has reached its peak. Further meaningful growth in that KPI likely comes from focusing attention on a different part of the experience, or testing a much larger change. 

Going back to the “emulate” approach can help - if there any obvious trends or new competitive features that have been launched in the intervening time you’ve been optimizing your experience. Again, the primary drawback to the “emulate then optimize” approach is that always you’re always behind or at parity with the competition. 

A needs-driven approach to roadmap planning and ideation can both set the team up for success in growing KPIs and create a differentiated product in the eyes of consumers. To accomplish this, some portion of the team (usually Product, Design, and/or Research) should keep up to date on what’s going on in the market, as well as in user’s heads. Activities like research studies, market research, customer segmentation, and user analysis should be ongoing and in parallel to the test-driven optimization efforts and insights shared at Learnings meetings, elevating that time spent together to a shared understanding of what’s resonating and working within the product today (test results), as well as what your customers’ needs are, and where the market is going. Look to adjacent industries. Transformative ideas often lurk in the spaces between “your business” and ones that are the same, but different. 

Understanding current product/test analytics, customer needs research, and market trend /conditions are not enough to break out of the local maxima, and are the key ingredients that the seed your team’s ideation efforts - design sprints, hackathons, etc. Without setting your context, ideation often spins wildly off into tangents and unproductive/unreasonable directions. Instead, arming a cross-functional team to ideate and test solutions for your user’s potential unmet needs in a defensible and  differentiated way.

The product managers who manage a diverse portfolio of projects that range from small optimizations and UI tweaks, to large scale changes to the experience driven by the research and ideation process are maximizing the probability of finding wins that drive their KPIs forward. Depending on how the current roadmap of tests are performing, resource allocation between this “portfolio” of product ideas can shift. If current test ideas are not working, it may be time to reboot the roadmap. If current projects are showing gains, a heavy focus on rapid execution of optimizations ensures that no wins are left off the table.