[tuckerconnelly]

From Peopleware: “In the 1985 Jeffery-Lawrence study [from the University of New South Wales]…they investigated the productivity of 24 projects for which no estimates were prepared at all. These projects far outperformed all the others…Projects on which the boss applied no schedule pressure whatsoever (‘Just wake me up when you’re done.’) had the highest productivity of all.”

I read 20+ books on management and leadership[1], and none of them mentioned anything like Scrum. I agree it's BS.

[1] https://tuckerconnelly.com/management-leadership

[jdougan]

While I too despise Scrum, the causation could be runming the other way: the Bosses that have a better team could be more likely to let them run without major pressure.

[intelVISA]

No love for Scrum but this is the more likely explanation.

A good team that runs itself? Ofc it doesn't need Ten Scrum Masters to deliver value.

Now, the real question is why leadership tries to salvage failing teams with Scrum? Save the wasted money, use it to hire top talent instead... easy.

[kmoser]

Top programming talent may still need to be managed (and in fact may need to be managed more than lower quality talent). The sweet spot is to hire top programming talent who is also good at intra-team communication and can organize themselves with little direction from management except to be informed about outside factors like client priorities.

Even better is top talent who can also interface directly with the client when necessary (not necessarily all the time) and doesn't need everything first filtered through a manager.

I've worked with teams like this in the past and it was always a pleasure. Most of them were fairly experienced devs and knew the value of email, phone calls, and water cooler talk (serendipitous discussions which led to valuable information being exchanged). Despite the lack of "modern" productivity tools like text messages, chat apps, and Slack, we were able to get stuff done efficiently.

We had weekly meetings which were productive and useful, and actually helped identify if anything was falling through the cracks. Nobody got bored because the meetings were actually helpful.

[fendy3002]

If CEO can hire a 100$ scrum master to save a 1000$ failing team, it's much cheaper than hiring 2000$ successful team. That's probably what they're thinking

[replyifuagree]

Sadly the $1000 team doesn't get saved by the $100 scrum master though.

You are 100% right that legacy management falls for the SCRUM sales pitch.

At one company I worked at the scrum salesman basically bullied the executive team by saying "You don't want to be the last company to adopt scrum do you?!!??!"

[kelnos]

I don't think that's true at all. If there's time-to-market pressure (for example), then you really want that $2000 team to deliver on time. "Saving" the $1000 failing team with a $100 scrum master may not be good enough.

Also, please at least acknowledge that your numbers are completely made up and may have no basis in reality. It might be a $1000 team only if they deliver on time, but the overages might push that to $5000. Or whatever. See, I can completely make up numbers to support my point too.

[fendy3002]

Saying that doesn't mean I agree with what I say, you know.

[kuchenbecker]

Also, this is needed now and needs to be repeatable.

[mdekkers]

> […] hire top talent […] easy

Have you _tried_ hiring good people?

[blq10]

Top talent does not exist.

This problem exists at big tech and startup, in companies that spend fractional multipliers of the average salary on engineers as well as those who pay poorly.

In this environment, if your solution is "hire better people', you can't- there isn't any

[epicureanideal]

> This problem exists at big tech and startup,

Big tech or startup status are not correlated with top talent. I've seen the whole spectrum of low to high talent from both places.

> in companies that spend fractional multipliers of the average salary on engineers as well as those who pay poorly

Pay and top skill are only slightly correlated, unfortunately.

> Top talent does not exist.

It does exist. I've seen it, but most managers don't know how to find it, identify it, and retain it.

[marcus_holmes]

The people who implement Scrum are insecure managers. They don't understand the development process, don't trust their staff to just get on with it, and need constant reassurance that their project/product is making progress.

So it would have to be both: the devs are good and don't need hand-holding, and the manager is able to deal with the lack of transparency that "it'll be done when it's done" comes with.

[koonsolo]

44 yo dev here. What I've seen from managers is that "letting go" is more of a personal trait than an external force.

Some people just have a very hard time letting go and trusting a team. Who they are managing just needs to follow.

[PartiallyTyped]

Our team is the only one not doing “scrum” or estimates or shit. Our team is the only one far ahead because we don’t waste our lead engineers’ time and allow them to move at their own pace (that means very fast). I just have 15 minute dailies with the other senior and the manager to stay in sync.

[guytv]

During a certain phase of my career, I was part of a company that extensively used the Scrum framework.

My takeaway then was that Scrum fosters a modular team management style, which diminishes the dependence on highly skilled individuals.

This approach seemed to offer management a sense of oversight in the software development process, but I didn't stay long enough to determine whether this was actual control and predictability or merely an illusion of it.

[baal80spam]

I re-read Peopleware this last weekend. It's one of my favourite books on the topic of software management. There are chapters that are not very useful for the $current_year (eg. regarding telephones or office furniture) but overall it's a fantastic piece on human interactions in the industry.

[SergeAx]

What if the boss gets a wake up call from bank, saying they are out of funds to pay team salaries, first?