Managing Performance With Escalation
Feb 16, 2018
~3 minute read

I’ve discussed with other managers the concept of having an escalation path for how you talk to reports about performance. In the ideal case, a manager should be providing feedback about possible concerns well before they become major problems, and then if something continues to develop into a larger problem, adjusting their messaging accordingly. The following is an example of how that might pan out across multiple (increasingly concerned) interactions.

  1. “How are things going with X?”
  2. “Do you need help with X?”
  3. “Hey, X seems like it’s been a bit slow, what’s going on there?”
  4. “We really need X to get done ASAP, please focus on that.”
  5. “Look, your work on X hasn’t been up to snuff, this is a problem. How can we fix it?”
  6. “If your work on X doesn’t improve it’s potentially going to have implications for your employment here.”

The first couple of interactions are less certain and more explorative. They’re the kinds of things that you might ask when you’re not even sure if there’s really a problem - perhaps your report just had an off week, or maybe you haven’t heard any news recently about the project and want to check in.

The middle two are when you have somewhat more solid concerns - there’s a specific thing that could be better, but it’s unclear whether your report recognizes that. They’re less exploratory and more directed, but still essentially requests - indicating your desires but still keeping the report mostly in the driver’s seat.

The last two are when you’re getting to “significant problem” territory. These should almost never be the first time you’re talking to your report about a particular concern; they’re for when something has been brought up in the past but didn’t improve. They’re not about exploration, they’re about information: letting the report know in no uncertain terms that something needs to change. The latter is basically a step short of a formal performance improvement plan (PIP), which should be the next step if something doesn’t change.

You don’t necessarily have to stop at every step of the path above - some serious problems might skip directly to 4 or 5 (or even 6, but that typically means something has gone very wrong). In other cases you might move faster along the path if something deteriorates quickly (perhaps skipping 2, 3, or 5). You might have your own in-between steps that you find work best for you. What matters most is ensuring you’re using the language that matches what’s really going on. Don’t hover around the earlier messaging if you’re actually about to run out of options.

« Back to index