The Dreaded Weekly Status Email
eleganthack.com
eleganthack.com
To be clear, I'm all for pushing stats to the central org. Every week when I was a company commander in the Marines, for example, we sent our battalion HQ a big Excel sheet with a summary of who was on light duty, who was going on leave, weapons count, etc -- that kind of basic accountability is fine to put in a weekly email IMO.
What is not fine is abdicating your responsibility as a leader by living within a Potemkin supervisory framework that reduces or even eliminates your need to have daily face-to-face (or telephone/VTC) conversations with the people you have the privilege of supervising in order to figure out what challenges are in their way and what you and the larger organization can do to help.
The straightforward way to do it is not weekly status emails, it's limiting the number of direct reports anyone has to a reasonable number and holding them accountable for everything their portion of the org does or fails to do. They in turn should be holding their people to a high standard of not letting bad news age and being forthright about what's not working.
The problem is that status emails containing substantive subjective summaries of the challenges faced by the person writing them are often used by incompetent managers as a replacement for constructive dialogue between them and their direct reports. At least, this was the case at the two companies I worked at that had a required status email as part of the weekly drumbeat. And I've heard a fair amount of anecdotal evidence from friends at other places that the problem is widespread.
- Brief summary of what was done since last report
- What's next
- Any blockers
- Anything else important (optional)
The only difference is the inclusion of OKRs/KPIs, which makes sense if you are using them. I tend to do something similar with project status reports as well, which will usually have red/amber/green in the subject line and if the status isn't green, the email starts with the reason!
Or sometimes you just really don't care.
Agreed. Hearing about the minutiae about someone's task is tedious. To me the daily scrum is like getting home and asking your significant other "how was work today?". Boring question, boring answer.
I definitely prefer the asynchronous approach. That way if I don't care about what you're working on (or vice versa), we don't need to waste time standing in a room together.
If you can't learn from what they're saying, don't need to know it, and can't contribute, then you shouldn't be there. This is why we use feature crews as a subset of the team.
If part of my work depends on theirs, or requires specific coordination, I'll use whatever mutually agreeable medium to exchange the information that is needed.
I've never really felt like I didn't know what people on the team were working on, despite not having a daily meeting or weekly status.
I don't get the blockers thing either. If I'm blocked, you're getting an email or IM immediately. I'm not going to wait until the weekly status email, or even the next day, to tell you. And I'm not going to bother the people who can't do anything about it.
But understand the hesitation. And if someone is blocked, agreed, it should result in a quick email or IM. Of course, if you spent all day yesterday trying to figure something out but are just feeling kind of stuck, standup call is a great place to say, "hmmm, I can keep working on this, but if someone has some time, I could sure use a second brain to help."
- Quick, 5 to 10 minutes
- Strictly focused on current status and blockers,
- Attended only by people who are directly working on the same project.
- Has someone designated to run the standup, who is empowered to table any discussions that start to drift.
If you don't have a tool or an easy to update document that shows this at a glance, then you have a problem. If you are asking people to give you all this information and then trying to collate it, then you have a problem. If people are not very simply causing this information to be visible as part of their normal workflow, then you have a problem.
Meetings are not the solution to any of these problems. What meeting are good for is: Human contact. People often find that if they don't interact directly with someone that the other person might have important information that isn't being shared. The longer the time between contact, the more that irrational feeling grows. Not everybody feels that way. I don't and it took me years before I realised that it is hugely important for other people.
The best thing I can say is that, if you don't have a short meeting where people can fist-bump, eventually someone on your team is likely to go crazy and cause huge problems. Usually that person is your manager, as it turns out.
My advice is to treat any substantive conversation in standup as being something to be fixed. You don't want that. You want people to show up, complain about how boring standup is and disappear. Every time you find important information showing up in standup, ask yourself how it could have been shared more easily as a normal part of the work flow. Keep working at it and eventually standup will be all fist-bump and no content.
You know what perpetuates people not taking the meeting seriously? Having a 10 minute meeting every day just for fist-bumps.
No affiliation, just seems like a cool product. Probably wouldn't be hard to whip something similar up if it's not already on GitHub.
There are a few others to try. I don't think any has a distinct advantage, just varying levels of polish. But I would recommend using it, especially for (semi-)remote teams.
I think you're right to wonder about alternatives, though. The aspects of Geekbot we use would be fairly easy to replace with a bot of our own. Geekbot pricing has a very narrow sweet spot, which I imagine varies significantly by team.
There is an ingraned urge to interrupt everyone and force people to waste 30 minutes of their daily time.
The concept is pretty similar. Not everyone uses it, but I like having it to keep track of my work.
(We work in teams of 4-6 and standup is never more than 10 minutes. I've also had the "30-minute standup" which not really a standup...)
That's why it was called stand-up, in order to emphasise that you stand while doing it so the meeting simply can't be long since everyone is psychologically incentivised to get to the point so we can all go to break room, sit down and work, or get coffee.
Of course it didn't work out that way in many places, and now 'stand-up' is a 1 hour long daily meeting complete with minute notes, rules of proceedings, and slide-shows.
Daily, repeated form-filling is hard for creatives. Never heard of Assembla but if you were my manager and you were repeatedly needling me to fill out some form, every day, when I'd rather just give you a few words over chat (or in person)... I would most definitely push back.
Of course, in my case, I didn't wake up on the first day and put this in place. I spent a few months trying to get people to send in their standups every day, but realised that I can't possibly do it everyday with an expanding team and it didn't make for the best use of my time. Because filling in your standup is a quick part (quick because it shouldn't take you longer than five to ten minutes to fill it up) of the process that needs to be followed, the most efficient alternative was to create a bot to remind those who can't remember to do. Plus, the number one reason I got from people who failed to do it was that they simply forgot.
Try trusting them instead.
Signed, Micromanagee
If I was telling them what to do, what results to expect _and_ how to do it, then I would be micro-managing.
Don't Lumbergh your way through it - "So, Peter, what's happening? Aahh, now, are you going to go ahead and have those TPS reports for us this afternoon?"
If the point of standups in scrum are to find out what an individual has been working on, every day, then this could feel like micromanaging to some people.
The end goal of standups, the way I understand it, is to figure out what people in your team are doing and what they are stuck at in order to:
* align people towards individual goals in the sprint that require collaboration; * remove impediments that are beyond the control of team members and that may cause delays; * improve task and load distribution, so that some members aren't overburdened while others are slacking off (if you run a short sprint and give team member autonomy to work, tasks that require collaboration from several team members can tend to become bottlenecks, as I have experienced).
I suppose you could well call it micro-managing if these standups were used to manage individuals instead of the process. In my view however, standups help you gather data to manage and improve the process, so as to make things favourable for the team.
If you have to closely observe or repeatedly needle your team for information, something is wrong. Programmers are largely introverts doing high-focus activities; we need to be left alone.
We (as in my team) don't strictly follow formal practice, though we have standups and I try to follow LEAN. Our git server sends email notifications with each push and I find these are wonderful for peering into what other people on my team are up to.
"In business management, micromanagement is a management style whereby a manager closely observes or controls the work of subordinates or employees."
If you replace the standup with written reports, I doubt anyone but the manager will actually read them. I know I wouldn't, regardless of how valuable it seems in the abstract.
In theory this sounds good, but that is an emotionally charged and negative way of saying it. "Tell me what you promised to do and if you actually did it." You would not use such a construct in a negotiation except to intimidate the other.
In certain people, this destroys motivation, even when they didn't achieve a prioritised task because they found a better way or reprioritised for good reasons.
If it was done, great. If it wasn't, do you understand why? Did you underestimate? Did something block you? Was it deprioritized? How will this not happen again, or why is it now okay?
If someone is less motivated by honest discussions about productivity, I'm not sure they can improve their productivity and collaborate with others long term.
I think if you badger employees about their estimates and not getting things done that they promised, that will lead to those employees overestimating the amount of time required, just to be sure that they don't end up not meeting the estimated timeline. That seems like a net loss for productivity. Rather than encouraging people to stretch themselves and get more done, you're leading them to believe that it's much more important not to miss self-imposed deadlines.
Estimating is important because it gives management the ability to deploy resources strategically and manage expectations. Bad managers think estimates are a promise/guarantee. Assuming the estimates were arrived at with as much clarity as possible (planning poker), good managers understand that estimates are a tool to aid decision-making.
For many companies, having a defined degree of confidence in estimates is a requirement of doing business. Any DoD contract in excess of $25 million USD, for example, must be performed by a company with a DoD certified Earned Value Management plan. You had better get your estimate process right there, because you're held accountable for it.
Typically the process is that the contractor has an internal pricing team that figures out how much to bid on a contract. After the contractor wins the bid (assuming fixed price), they then staff it with some combination of employees with various rates to meet the bid cost. Then those employees write up a detailed set of requirements (keep in mind that this occurs after the bid has been accepted), and then the project begins and things really get screwed up. Life happens. People leave the company. Clients are waaaaaaaaaay slower than anticipated taking care of required administrative work like approving user access and granting PIV cards. Clients change their mind or didn't account for several important considerations that may or may not involve huge amounts of work.
So how do the companies remain profitable through all this? It's usually that as the due date moves closer and closer, employees are expected to just work more hours to account for the missed estimate.
This may come off as cynical, but the truth is that there is not a soul on Earth who could accurately estimate how much time will be required on a multiyear contract given how many variables are involved. So, I'm back to my original point- estimates are largely worthless and a waste of time. Yes, they are a means to an end, but are in no way a reliable indicator as to how much time or effort is actually required to complete the work.
As an employee of a huge contractor myself, you aren't wrong.
The contracts people will always want to see the EVM plan. The contractor's budget people will compare the estimate to prior actuals and somehow come up with an estimate they think is good. And then you will run over, and the government will try to hold you accountable, even though the causes of your overruns may be requirements changes from the customers.
I apologize if my original post presented this rosy picture that estimates are wonderful and always work. I guess some recent training came through a little too much.
My experience has been that the only questions management should bother with are usually the issues that managers can actually fix. That tends to be a surprisingly limited set of things.
That is my point. If you want openness from people the original phrasing might not be the best. Nailing them with questions won't help either.
If a manager is not able to take into account human nature and it's quirks, they certainly can't improve anybody's productivity long term.
It doesn't sound good to me, even in theory. If I'm becoming aware of what is or is not getting done on my teams via an email once weekly, I'm really not doing my job. I should have a general sense of where things are at a more granular level than that, and be able to synthesize that information through the regular communications of my teams so as to not impose administrative burden (which takes time away from their work). Assuming I'm doing that, I can be the one that writes about it if someone else really wants that from me. That's part of my managerial responsibility to keep shitwork away from my folks.
More like, "Give me one good reason why I shouldn't fire you right here and right now. Ok, well then, give me another."
The only time I asked members of my team to write them was when I needed certain individuals to really think about their work and focus on what was important vs low priority. We talked about it, but then I had them write status each week for a while mostly to force them to stop and think about the connection of their work to the larger business objectives. I really value developers who have some level of business acumen and ability to understand the "why" of their work at a more strategic level - and if someone doesn't have that I'm going to coach it. That's just one way that works for some people. More of a coaching tool than an actual status report, though.
1. What did you focus on this week?
2. What are your plans and priorities for next week?
3. What challenges or roadblocks do you need help with?
4. Is there anything else on your mind you'd like to share?
Managers then can review and comment on each part of the check-in. That happens the same day, which means that it doesn't feel like a waste of time. Even just an emoji as a comment can provide positive feedback.
I like this process because it's efficient. I also enjoy a log of what I have done in past weeks to reinforce that we are progressing. Lattice is primarily a goal-tracking app, so tying goal updates to weekly check-ins makes the process easier.
This works for us, but we're a small team. Weekly status emails sound like a different beast relating to tiers of tiers of teams.
I suspect that part of the difficulty of the "weekly status email" is keeping track of updates from memory. As the author describes the Zynga framework, it's clear that it's easy because it asks for fairly objective things - like OKRs. It sounds like scaling a solution like Lattice to larger companies would ideally prepopulate reports with data - like "Closed these 7 goals . . . , closed these 8 pull requests . . . , analytics shows goal X did Y" - then allow the manager to edit and provide context.
Are there effective teams who don't do anything like this at all?
That's a process problem. Teams I was on with this problem needed to get better at spreading such background knowledge (e.g. by having code review).
I'm pretty sure that the only effective teams are the ones who don't do anything like this at all.
Perhaps there are projects that require this level of management and communication, and perhaps there are people that work well with it. But there are also probably many projects that do not need such management overhead and where some types of people do not work very efficiently in. So, what I'm saying is that I think this management style shouldn't be an assumed default.
You can guess what happened next -- someone would check the document out of Sharepoint, locking it for everyone else. Then that person would inevitably get interrupted and go handle the latest crisis. Meanwhile everyone else was unable to enter their status into the shared Word document.
Eventually management gave up and returned to having us send emails.
Fortunately I don't work there anymore.
Would they have/did they bother to go through the Word document either? To me, it seems like a fun new way to test a Markov chain generator.
What about reviewing the data for long-term trends, rather than gut feelings?
I do feel like OKR has a fair amount in common with GTD as well, though I'm not actually familiar enough with GTD to assert that, though from the outside they are similar.
But joking aside, plans that rely on trusting everyone to act in a certain unwritten way aren't very reliable. It's like how one study showed that mistakes during surgery went down by a significant degree if doctors simply compiled a checklist of what they were going to do beforehand---even if they had done the procedure hundreds of times before.
People, in the middle of working on hard tasks are not very reliable. That's not to say they're not smart, or professional, or lack care for their coworkers. It's just a fact of human nature that we solve problems in whatever organically directed way makes sense at the time, and not what the 'best over many averages' way would be.
The checklist is only there as a public contract that "if I don't do X, then anyone who points that out is doing the right thing".
This is both more frightening, and more insightful, than the "people forget to do stuff" lesson (though that is true too).
http://beyondthechecklist.com/2013/03/thank-you-for-pointing...
If you're a manager, then the reason they do that is because you intimidate them. They're afraid to tell you when something goes wrong. This is YOUR problem.
plans that rely on trusting everyone to act in a certain unwritten way aren't very reliable.
That's the only possible plan that can work, because you can't write everything down. At some point you have to trust.
But no, I'm talking about people that I have worked with, but not managed.
And these aren't 'dysfunctional' teams by any objective measure. These teams were formed from reasonably competent individuals, typically turned in their work on time, and had comfortable social dynamics.
Yet on the rare occasion that a problem cropped up.... ah we didn't always take the best approach to problem solving.
----
There's a story I heard about firefighters. They did a study on veterans to find out how people operated under pressure. The researchers thought in the face of a burning building, surely the fire-fighters wouldn't be able to do an in depth search of all possible solutions, so they'd just compare the top-two that they thought of.
Seems reasonable.
In reality, the they just went with the very first solution that they didn't imagine had fatal errors.
That's how people are... under pressure, we go with the first thing that might possibly work, not the best thing. That's why we need to think about problem solving techniques, before problems actually occur.
Sure, "No plan survives contact with the enemy", but I've always found the process of planning to be invaluable.
As others are noting, all this is very subject to abuse, but it can be a very good thing.