The Zappos holacracy, Edward Tufte’s sparklines, and an 11×17 printer | Hacker News Reader

The Zappos holacracy, Edward Tufte’s sparklines, and an 11×17 printer

blogs.law.harvard.edu

31 pointsjessaustin25 comments

Having spent multiple decades in big technology corporations (now out of that and in startup world), one thing I started noticing towards the end is the dilution of power of Execs’ due to information becoming freely available to all and sundry in an organization.

These are strategic level information (the technology trend, shifts, competitive landscape etc.) to information on company metrics which are now freely available publicly and it can be argued that even Exces’ would not be aware of all of these info (blogs, analysis, leaks and even things like anonymous public forums where employees speak out - not limited to ‘Glassdoor’ and its ilk).

Few decades back none of these information were available to an average joe employee, they were accessible only to Execs - strategical knowledge through paid researches, consultants, high paid industry advisors etc.

At that time information WAS power, Execs’ would take decisions and employees would follow, they had no room for doubts as they did not have any information with them which would make them to think critically. However, now things have changed so drastically that today a young employee is much more clued in to technology trends, shifts and competition than an old world clueless execs’. If you throw in jaded middle management in this mix, the scenario gets even more ugly.

I guess this liberation and accessibility of information is going to play a very big role in coming decades as to how organizations gets shaped.

Do I really want a piece of paper? Or a dashboard with the same contents - perhaps set as my browser's home page. With the addition of a few large format screens showing the dashboards in the work area, I think that would suit our group better than a printed piece of paper.

I think the more important concept is to get everyone in the group focused on the core metrics and one thing I do like about the idea is that there can be more than a "few key metrics".

I'm imagining every one of our application servers counting the errors that show up in the logs - right next to the raw hits and response time. We're responsible for more than 80 products and it would also be informative to show the error rates of products versus each other. Which of the products needs maintenance work most?

We capture metrics for some of the newer applications in Ganglia, but the idea that you might fit 500 numbers and spark-lines on a single dashboard has given me so great ideas.

> Do I really want a piece of paper? Or a dashboard with the same contents - perhaps set as my browser's home page.

The piece of paper would be sitting there, waiting to be seen. When's the last time you looked at your browser's home page? I don't even think mine has one.

And changes to a dashboard-generating system can be a lot more expensive that changes made in a paper-printing system. As someone elsethread noted, he's seen millions spent on tools to replace cents spent on paper.

A project management dashboard I have been trying to adapt to my work is the "One Page Project Manager". It is a sort of simplified Gantt chart and RACI.

http://www.slideshare.net/mobile/GeoffatPerformancePeople/us...

https://oppmi.com/

The idea that a company would have 500 meaningful and actionable metrics that would each produce relevant data daily seems laughable.

I've always found that the more time, money, and employee hours a company spent on tracking metrics, the crappier that company was. (My sample size here is admittedly small.)

I've usually found that when metrics become the focus of an organization, the organization optimizes around gaming the metrics. This happens in companies, volunteer groups, communist dictatorships, capitalist paradises, everywhere.

Something inevitably gives.

I also have a hard time believing that this idea is floating putting a piece of paper on everybody's desks, in 2015.

An investigation of UK, specifically English, education policy and practice over the last decade or so would provide evidence for your position. The thing that gives is time spent in dialogue with individual students.

Paper? We have dashboards. And colour printing is a bit expensive.

I use paper all of the time to present information. It's remarkably effective and easy to do vs. screwing around with dashboard apps, accessing various data sources, etc.

To avoid the horrific costs of printing things on color, I've seen millions wasted on BI tools and other fluffery. 10 minutes of excel and 10 pieces of paper does the job.

That is only a single paper though. Now you must replicate and distribute that paper to every person in the company...ASAP.

The consumer of metrics that I give a hoot about are a few key team members, the COO and CTO. I go to "# of copies" hit 6, and I'm done.

I'm not a total luddite btw, Most of what I do is project based or operations in transition. So metrics are either evolving or we don't have a consensus over what matters. So the building process of automated dashboard isn't a high roi activity.

This is one of the tricks of metrics. They are made to be gamed. So, in picking the metrics to measure, you are picking the game you want your team to play. Trying to cover all metrics is silly, as you context switch between too many things.

If you can't measure what's important, then what you do measure becomes important

What gets measured gets managed. - Peter Drucker.

This idea sounds awesome, but what is this 'printer' or 'paper' that he speaks of? Is that kind of like a paginated web app?

Agreed, it's impossible to make decisions on 500 metrics. The art is focusing on a handful of metrics that matter most.

No one would consider 500 metrics on a regular basis. In the "holacratic" context, however, it might be valuable for the worker to be able to consider different metrics today than she considered last week. It might be difficult to decide ahead of time which metrics will be useful. With too few metrics, that very decision might undermine holacracy.

There's a large difference between having access to metrics/data and an operational dashboard. If an employee is curious or wants to prove a hypothesis, the data can be made available through tools like Chartio, Looker, Periscope, etc. But a dashboard should represent the focus of the business and team. If she is looking at different metrics each week to make decisions, one has to question the integrity of her focus and decision making ability. If you're going to distribute metrics to employees every day, they should represent the current business goals and clearly communicate the businesses progress.

For anybody else that was confused: 11x17 = A3 (this might save you 30 seconds of googling)

11x17 at least gives a sense of dimensions (provided you recognize those as a standard imperial paper size). With the A series of paper sizes, most people know A4, then don't remember if sizes go up or down numerically.

Not the case in the UK, I suspect, which is where I live - most people will be familiar with A3/A4/A5. They've been the standard for the past 50-odd years, and penetration of non-ISO paper sizes is approximately 0%.

Until I followed the link, I didn't know 11x17 wasn't just a particular type of printer. And once I found out it was a paper size, I wondered how large it was, since I don't have a good feel for anything in inches once you get past the 12 that fit on a standard 30cm ruler. Which is why I posted :)

> Not the case in the UK, I suspect, which is where I live - most people will be familiar with A3/A4/A5.

Exactly. A4 is for letters and documents, so it's everywhere, people know A5 from their exercise books at school, and will have used A3 for something at school, perhaps art or technical drawing.

Beyond that, flipcharts are A1, and postcards are A6, although i suspect most people don't realise that. You don't see A2 much.

I once came across a giant A0 flipchart pad, and immediately attempted to make a paper aeroplane from it. However, it's too thin relative to its size, so the plane collapsed.

This is one of those USA-vs-everyone-else things.

If you grew up in the USA you're used to quoting paper size in inches.

If you grew up anywhere else, you're familiar with at least A4 and A3 sizes from the ISO 216 standard.

In case you're wondering, ISO 216 series of paper sizes begins with A0, which is exactly 1 square metre in area with an aspect ratio of 1:√2.

The square root of 2 comes about because you produce A1 by cutting or folding A0 along the longest edge. A2 is produced by halving A1, A3 by halving A2 and so on down to A8.

> This is one of those USA-vs-everyone-else things.

I don't think it's quite that simple. I was recently with a European friend who had ordered some prints, and she told the front counter she'd ordered A3 size. She knew that the prints she wanted were much smaller than a regular A4 page, and of course 3 is a smaller number than 4, so logically her small prints must be 'A3' size...

This headline sounds like it was generated by a Markov chain reading hacker news.

"What do readers who work in companies with at least 50 employees think? Would it be helpful to arrive at work and find an 11×17 sheet of paper across your keyboard with perhaps 150 metrics of the company’s performance?"

Performance metrics like these don't really help with many of the kinds of decisions we pay high-level managers for, especially long-term strategic planning. For example:

- Do we want to move into new line of business X? Would purchasing Company Y help us do this?

- What qualities are we looking for in the person we'll hire to replace our CTO, who just turned in her resignation?

- Should we be trying to negotiate a better deal with our health insurance provider when the contract renews?

- Should we change the way we recruit developers? Should we open a new office in Shanghai to get access to the pool of developers there?

- Is government regulation in our industry likely to hurt our business over the next few years?

The company still needs people to deal with questions like these, and real-time operational data doesn't really help with these kinds of problems.

One thing that essentially every organization has is specialization (job roles). When I was working as an engineer, my problems focused around engineering. I simply didn't have the time to think about company-level problems every day.

Reply on news.ycombinator.com