Have your team predict your KPI's in real-time so you can actually meet your KPI's

By Adam Siegel on February 28, 2017

As a small company, we're constantly worrying about our next stage of growth. And like anyone else, we have a few "key performance indicators" we track to be aware of how we're doing in the things that influence our revenue, profit, and the culture we want to strive for. 

But while retrospectively tracking our KPI's has been informative, it is less helpful in actually informing our decisions. Ideally it would be nice to have a better indication of how we're going to do on our KPI's weeks or months in advance, to increase our chances of actually meeting them. 

Up until now, we've all just been...talking about them, but I wanted to see if we could bring some structure and feedback to this process. After all, we sell prediction market software to some of the largest companies in the world to do exactly this, so even with our small number of people, we should be able to eat our own dog food.

Recently, we launched TinyCast, a free prediction market to "Ask your team - the people who know best - to predict the future. Crowdsource the odds and get essential insider insight into your most important decisions." So instead of setting up a whole private site for our company, I've just been asking a series of TInyCasts.

We've always been pretty transparent with our team about how we're doing on our numbers, but launching these TinyCasts we've already seen a few benefits:

  • Everyone in the company, from the engineers to the designers, are becoming aware of what's going on with the business
  • We're uncovering people's true beliefs. We can sit in Friday update meetings and everyone nods along when I talk - but what are they really thinking?
  • A reality check for my own biases as the co-founder (and co-biggest fan) of the company and what we're doing

Here's an example of a TinyCast we're running and how I set it up:

Within the next couple months, we're going to be launching a new self-service business model for our internal crowdfunding product. Up until now, this has always just been an "enterprise" product for us: people have to call us up, we have been wrapping professional services around it, etc. But I'm a big believer that the addressable market for this product is quite large and I've been pushing us to get the application to the point where we can put a registration/pricing page in front of the application, have a simple on-boarding process, etc. to really try and scale it. 

But how does everyone - the people who will actually be building, designing, and marketing this capability - actually think this will go? So I, *gulp,* created a TinyCast to see:

It's early days, but the leading signal is "11-15." A little lower than I'd like but this isn't about what I'd like to happen, it's what we think is actually going to happen. As time goes on, our team can come in and change their forecasts at anytime (as you can see on the trend graph is already happening.) We can also all collaborate in the discussion thread on competitive activity, product dev, and marketing updates to inform each other and keep our thinking current. 

Besides being able to get a raw data dump of all the predictions, I also plan to generate this nifty pdf I can bring to our Friday meetings to discuss what the latest predictions are. 

Frankly it's a little scary to ask this and some of the other TinyCasts I have out there for us. It certainly flies in the face of typical corporate culture which is explicitly to not be this transparent. You just don't know what you're going to get back and you could quickly find you have little support for what you're doing. But ultimately I decided I'd rather know early so I can either explain why we're going to stay the course despite a lack of confidence in the outcome, or why I've been persuaded to change course based on these results. 

Long term if we run enough of these, we can measure how accurate people are, where we have biases, and ideally become better and better about forecasting our KPI's but first I just want to focus on having people respond to these and update their forecasts on a regular basis.

I've always liked to believe our culture is such that people will just tell me if they believe in what we're doing or they think I/we are full of shit, but short of that, the numbers won't lie.

Not thoroughly bored, possibly even intrigued by this post? Then you should follow us on Twitter @Cultivatelabs

prediction markets disruptive leadership crowdsourced forecasting