Choosing a metric for a skills management platform
Through the summer, we have been hard at work on the design of TeamFit v2. Our users have helped us to rethink some of our earlier design decisions. Users tend to do that. We will post more on these users driven insights in a couple of weeks; today, the focus in on metrics. Partway through the design process, we realized we had to understand what metrics we wanted to optimize and use this to guide our design decisions.
A lot of our thinking on metrics comes from Alistair Croll and Ben Yoskovitz wonderful book Lean Analytics. In this book, they emphasize on the importance for early-stage companies to focus on ‘the one metric that matters.’ This makes a lot of sense to us. Having one metric focuses the team and makes it easier to communicate our progress to investors. As Alistair and Ben point out, focusing on one metric helps the company in four ways (see page 58).
It answers the most important question you have at any one time.
It forces you to draw a line in the sand.
It focuses the entire company.
It inspires a culture of innovation.
Lean Analytics suggests that your single metric that matters should be the one that tracks the greatest area of risk for your company.
But what is that?
We came at this from three angles: engagement, value and the core transaction.
With TeamFit v1, we were not happy with our engagement metrics. Several years ago, a partner at Charles River Venture Partners told me in regard to another company that “the best leading indicator of the success of a B2B SaaS company is user engagement.” This is a pretty compelling statement and represents a real shift in the world. Let’s face it, use of most enterprise systems is coerced; very few sales people actually want to enter data into a CRM but they will if their commissions depend on it. We have also been doing a lot of thinking on how to take all of the data collected on TeamFit to come up with a metric for what we call Predictive e. This is a composite measure of engagement that can actually predict future engagement. It looks at all the different user interactions and runs a predictive analytics system to understand what current user actions predict future actions. There are a lot of user actions on TeamFit that could predict engagement: claiming skills, providing skills evidence, joining project, suggesting skills to other people, endorsing other people’s skills. Somewhere in here is our magic number. The big social media platforms seem to be good at this. They know what actions lead to adoption and guide people towards those actions.
But engagement itself does not necessarily mean value. Over the long run, it is value that will get people to use TeamFit and keep coming back. TeamFit has to create value for three distinct groups of people: for consultants with skills who are looking for work on projects; for allocators, resource managers and project managers trying to staff projects; and, for company leaders who need to understand the real capabilities of their organizations and how these connect to the ability to deliver.
So we asked if value metrics for one or the other of these groups, or for all, would be the best thing to optimize. We definitely believe that value optimization for these three groups of users is our design goal. In our design thinking, work we made sure that there were ways to draw a direct line between the user experience and value creation. We are still boiling down the value metrics but we think they are something like this.
Consultant → Have a high utilization ratio
Allocator → Staff projects with high quality teams and manage the utilization ratio
Executive → Invest in the capabilities (skills) that are driving growth
We can design to capture any of these metrics (and we are) but none of them seem to cut to the essence of what we need to focus on or to capture the critical uncertainty about the platform.
Platform is the key word here. TeamFit is a platform in the sense that it is a two-sided market. We bring together people with skills so that they can form teams and work on projects. The people with skills are one side of the market. The people who need to staff and deliver projects are the other side of the market.
This is a simple representation of TeamFit but it gets to the essence of what are doing. Putting together people with skills as teams to deliver projects. In their book, Platform Revolution , Parker, Alstyne and Choudary emphasize the importance of the ‘core transaction’ for early-stage platforms. The core transaction is the nexus where value is created for each side of the market, where the two sides find each other, exchange information, and decide (or not) to do something together.
For TeamFit the core transaction is People Joining Projects.
That is what we have settled on as our one metric that matters. TeamFit v2 is being designed to optimize for People Joining Projects. We are now mapping the paths through the system that lead to this outcome and how we make sure that it is easy for each side of the market (basically Consultants and the people building project teams) to provide the necessary information to find each other and to build high performing teams.
Top image of Alex Calder’s Blue and Yellow Sickles (1960) courtesy of the Gagosian Gallery. Calder’s mobiles and stabiles are a good metaphor for the balancing act involved in building project teams. And the elegant way in which they move reminds us that teams are always in motion.
