When I speak with VPEs and other technology leaders, I ask them what metrics they track to for their exec teams, and their answers are often either thin (they’re tracking only high-level qualitative output like what quarter something is going to be done) or non-existent (they’re really not tracking anything at all). Most often, their reasons are some combination of a) the data is difficult to produce, and worse b) no one is asking for it or knows what it would be. However, even putting aside all the well-documented benefits of data-driven engineering, there are just too many reasons why a metrics-light (or gut feel) approach to engineering management and reporting can backfire for VPEs NOT to measure. This is especially true when times are tough. The biggest thing you need to understand is that if you don’t make the effort to measure, others will do it for you, and you probably won’t like what they come up with.
Engineering Output Quality
Let’s start with the easiest of the bunch, the quality of the work done by you and your organization. That is, recognizing quality can mean many things to many people so instead just measure the input/outputs of quality. Some of the most common complaints by sales and other leadership include the product is taking longer than expected to develop, or that it’s too buggy, or that bugs aren’t being fixed fast enough, or that… If you want to stave that off, you need to get out in front of it.
Take the time to pull metrics not only the new work being done but also the amount and cost of work to support your products. If you don’t get in front and drive the conversation, others will draw conclusions for themselves. I often talk about the allegory in which the sales guy doesn’t bother to show her funnel all quarter because no one asks for it. Then after missing the quarter she finally shows her thin pipeline. Too little, too late. No one trusts her anymore. Don’t be like that.
The same is true of engineering. Show your eye is on the ball. Monitor uptime, support tickets, bugs reported, and time to resolution, and what the team is doing to improve these engineering metrics.
Many companies I meet want to at some point try to put incentive structures into place for engineering. The reality is, for smaller, growing companies who haven’t had the conversation yet, it’s not a matter of if, but when someone will try. As a general rule, companies will always want to look for ways to measure the performance of their employees for bonuses, promotions, or performance reviews (and yes, given the current climate, layoffs). At some point, if engineering leaders don’t have ways to measure the performance of their teams, the company will probably assign one, and it likely won’t be the right way to assess your team.
Budgeting and Product Planning
Planning happens with your input or without. When it comes to the roadmap and execution, if you don’t build a compelling plan, back it up with numbers, show your work (best case, worst case, likely reality), you may find a plan is hoisted upon you. This includes not just the roadmap, but also timeline and budget planning. At the end of the day, you are in the best position to make realistic plans. Show the management team and remain in the driver’s seat.
Product Deprecation & Engineering Capacity
Lastly, many companies find themselves supporting too many products and features because leaders can’t make the case to deprecate the old and invest in the new (or vise versa!). Remember that the business may not fully understand the costs to support legacy products. Ultimately, supporting old products will get in the way of pushing new products out the door. It’s a zero-sum game. Illustrate just how much you are spending on supporting old products. Drive the business to have these hard conversations and to make difficult decisions.
Engineering leadership is a tough enough job without others putting out metrics, plans, and narratives for you. Get ahead of it. The idea that metrics are too hard to generate and aggregate is an antiquated one thanks to tools like Engineering Management Platforms among others. Do yourself a favor, control your team’s destiny because you’re the best person to do it, and put engineering metrics together, because if you don’t, someone will do it for you.