Why Reporting Sucks


I don't know a single person who likes reporting. Seriously. I never met a person who said: I like reporting. I know a lot of people who demand data for reports from others. But even these people hate it when they have to do it themselves. Reporting comes in different colors and shapes. I think we all experienced reporting by Excel throughout our career. You are asked to fill out some cells with values by a certain due date. It might also be a more informal status report during a team meeting. Or maybe during a daily scrum when you answer the question "What are you working on?" - while this isn't exactly meant to be reporting in the classic sense it is still some kind of status sharing.

Hidden intentions of reports

It's worth spending a thought on why reporting sucks. Reporting can be powerful if done right. There's a ton of tools out there which can be used to get better using reports. So why do we often seem to fail here? Maybe we can all use the insights gained to make reporting less sucking in the future.

In the past I had to do a lot of reporting myself. Here's an overview on what I experienced throughout my career (not necessarily in the companies I've been working for)

  1. Status Reports

    Status reports are the kind of reports I had to deal with most often. The goal is that somebody wants to gain insights in the status of a project. So maybe my project lead asks for a number of hours left on a certain task. Or my boss asks what the progress is on a certain job. This kind of reporting can help to bring more transparency into your deeds. And after all this is the core idea of reports.

  2. "Cover your ass" Reports

    These kinds of reports are never really called "Cover your ass" reports. Instead they are covered as Status reports. However, in this case reporting is executed "post mortem". So all the number crunching is made after the jobs were done. In result the numbers are more a proof of work. Often these kinds of reports are demanded by managers to be able to take the numbers and pass it on to the next level of management. I imagine those numbers being passed with the words "See - here's all the work we did in this are. You don't want to call us lazy, do you?" This kind of report guards you from being offended and from becoming vulnerable.

  3. "Process driven" Reports

    I've seen multiple cases where numbers are being reported because the "process" demands it. Nobody really knows why a certain field has to be filled, but everybody does it and that handbook about how to manage projects successfully claims that you can build a critical performance index out of that number. So better report it.

  4. "Just do it" Reports

    I'm sure you've seen those as well. Someone asks for a number and you have to report it. If you ask why, whoever asked you for this specific number just answers something like "Well, I actually don't need the number but XYZ asked for it, so we better give them these numbers. Don't question it!"

  5. Show-off Reports

    When the true reason for reporting isn't obvious, people start adjusting the message in a report to become heroes. I've seen this a dozen times and you probably have as well. Not sure what to put in that field and what the exact criteria is for a count? Well, let's just put in the maximum, because it looks better.

  6. Scapegoat Reports

    This last case of reporting is something I crossed several times in my career and I came up with the term "Scapegoat reporting". When something goes wrong people in charge might try to shift responsibility for the mess by reporting it. However the report is not made in a traditional transparent way to make people aware of the problem but instead it is hidden in a way that a) the person in charge can easily say: "I reported it and made the problem transparent" while b) the person who is in charge to recognize the problem will most likely not recognize it.
    This might be in form of a very lengthy email where the problem is not clearly pointed out but phrased. In case of an email the person reporting has even written evidence. It will be hard to acuse him of not being transparent later.

There might be other's but I think that's enough for now. It's obvious that except for the status report all other kinds of reports are indicators that something is going wrong in your company. Still I think the status report should - theoretically - be strong enough to make all other reports superfluous.

So why do we get bored by a fully legitimate status report my boss demands from time to time? Why does it seem so hard for us to put numbers in fields once a day, once in a week, once a month? Why does the time invested in reporting feel like 3 hours when it really rather 1-30 minutes?

Why it's not the tool

From what I found myself thinking and what I heard from colleagues and co-workers the first conclusion you might draw is that it's a tooling problem. The reporting tool might be too slow, too big, too confusing, too complicated, not integrated enough, too integrated, wrongly scoped. I know people complain about bad tooling a lot and I think I'm one of them. Sure, it is a tremendous waste of time if the tool you are using recurrently is designed badly. While this certainly could be a problem I think, this is not the root cause of frustration when it comes to reporting.

No effect? Skip the data.

The true problem is that status reporting - and therefore the kind of reporting that is being done with the right intension and should help you getting better- lacks effect because no matter what is being reported nothing happens and no actions are started as a result of the reporting.

Example: You might know the situation where people are asked for a status in a meeting. People point out that difficulties are rising and that something might be delayed. What's the consequence? There's none. Another example: You put numbers in a list. Sometimes they are smaller, sometimes they are bigger. Sometimes you forget about it. What's the consequence? There's none. In an ideal world people would use the data being reported to find out where things might go wrong in the near future and would try to avoid this by countermeasures. Like: "Oh - I see a decline in number XYZ. Let's discuss how we can bring this back on track, maybe we can ask ABC for help here. "In reality unfortunately often nothing happens until it's too late. Reporting is basically reduced to the action of collecting data instead of working interpreting and working with data.

Basically, this means that the lack of action based on reported data leads to a lack of trust, recognition and appraisal of reporting. Reporting becomes a meaningless activity which isn't here to help in the sense of true "status reporting" but to fulfill one of the other effects mentioned above. Clearly this is something nobody really wants and so people don't see a value in reporting and start to hate it. The reported data becomes worse because people try to minimize efforts and as a consequence the activities deviated from the reports become even worse as well. And here you are, right in the middle of a vicious circle.

It all comes down to some basics I already pointed out in my article "Thoughts on agility": While all we do should be commonsense, somehow we manage to behave like idiots. It's our job to make sure we break though the insanity we experience when it comes to reporting. Reporting is here to show what works well and what doesn't. To provide a prediction of the future based on facts collected in the past. We have to carefully scope and define the data we want to see in a report so that it really helps us to learn and improve our own process. A "collect it all" mentality is really only acceptable if you don't do it manually and if you know how to handle all the data afterwards so that you can use this data to learn and get better.

If the data we are delivering for reports don't seem to have any effect, it's our job to question the report. It's part of the "always learning and always improving" mentality every modern company should follow. Otherwise it's just a waste of time and resources.

Comments (0)

Skip to main content