Ad-hoc status reports considered harmful

Managers can’t help asking for status reports. It’s just one of the things we have to do when we’re asked a question we can’t answer with our brains or the tools at hand. We usually don’t think of them as status reports, we just ask a question like “Hey, Pat, when is Project X going to ship?” And that’s when the trouble begins.

It turns out, asking for an unexpected status update for a project immediately increases the perceived priority of that project. When you do this, there’s a good chance you just scrambled the recipient’s todo list for as long as it takes them to get you the answer. And probably longer.

I learned this lesson from Dave Loftesness while managing engineers at Twitter.

The story Dave told was a familiar one, in his own words:

When I was running the Search UI team at Amazon, my boss had tasked us with three major priorities, one of which was price sorting. Turns out, it’s tricky to sort by price when an item might have several prices: Amazon’s price, a partner merchant’s price, a 3rd party seller’s price for a used copy of the item, etc. The team and I decided that the other priorities had greater bang for the buck, so we put price sorting aside for a time and focused on other higher-value projects.

Fast forward several months. Jeff Bezos, as he often did, forwarded a customer complaint about our shoddy price sorting to my SVP simply asking “what’s up with price sorting?” Nothing else. The SVP responded “on it”, then forwarded the email to my boss, with a small addition: “we’re on this, right?” My boss sent a brief reply saying “oh yeah, dave’s been leading up an effort to clean this up. where are we on this, dave?”

Uh oh. At each step, the message escalated slightly. By the time it got to me, I was sure we needed to scramble the jets and get the eng team to whip up a prototype of price sorting over the weekend. Luckily, after some late-night consultation with my tech leads, I realized this was impossible due to the complexities that originally led us to focus on other tasks. So I swallowed hard, wrote up a detailed account of why we hadn’t made progress and had prioritized other projects higher, and gave a rough sense of when we’d be able to tackle it properly. Thinking back now, I probably bumped it up a couple notches in our stack ranking due to this inquiry from my management chain…

You see, when your boss comes to you and asks “when is X shipping?” it’s hard not to assume that someone higher up has been asking them the same question (even if you don’t have the benefit of an email chain like Dave’s example). And at some companies that may not be a good thing. Even if you don’t think this consciously, you are now thinking about this project, even if you had higher priorities, and wondering whether you’re about to be called to task.

This effect seems to be a common human bias. With awareness, perhaps its effect can be reduced, but we need also need to work around it.

  • First, ask yourself, do you really need to ask right now? Are you just being nosey or curious? If it’s Saturday morning, keep in mind that you might be ruining someone’s weekend.
  • Is there some way you can find out on your own? The team may have an updated project plan that you can just read. Making your reports do work you can easily do yourself isn’t delegation. It’s just laziness.
  • If you really do need to know now, ask yourself: why don’t you already?. Are you missing a communication channel? Is there an existing channel not being used?
  • If you have good reason to ask, at least try to ask better.

Here’s a bad example:

Hey Pat, when is Project X going to ship?

This could very easily cause Pat to get neck deep in working on Project X, even if there are other established priorities. Worse, Pat may not have complete information about Project X, which could lead to forwarding the inquiry to the team and generating a cascade of work and status reporting.

Better:

Hey Pat, I’m putting together a Board deck and need an estimated ship date for Project X. Deck is due Thursday, so if you can get this date to me by EOD Tuesday would be great. Also, why don’t we update estimated project ship dates on the wiki each Friday?

Pat may still get sucked into Project X, but hopefully we’ve reduced the chances of that and at least provided context on why the information is needed and how it will be used.

And for the recipients of such an inquiry, here’s a suggested response:

Sure thing. I’m booked up on Monday, but have a 30 minute slot free on Tuesday AM where I can work on this instead of prepping for our 1-on-1 on Wednesday. Hopefully that priority change is correct and a 30 minute timebox is appropriate. Let me know if you think otherwise! Looking forward to discussing your wiki idea at our 1-on-1.

Managing people is complicated! Sometimes the obvious approaches don’t work well. Or they work good enough to get by, but leave some collateral damage in their wake.

Thanks to Dave for both teaching me this and helping write this piece.

Software engineer at @czi-technology.