Blog posts on data

Posts selected fromManagement Blog - Engineering Blog - Investing Blog and other blogs

• Understanding Data

  Effective use of data is an important part of management improvement. Emphasis is needed on “effective”; data misuse is rampant and creates problems. Use of data by itself is not sufficient.

  To be effective you need to learn to think about not what is printed on the page but what lies behind the numbers you see. The numbers are just proxies for the real situation.

  continue reading: Understanding Data
• The Customer Knows Best?

  Some management ideas are pretty easy and straight forward. But many management practices require knowledge and judgment to apply them successfully. Easy solutions may be desired, but, often you must choose between easy and effective (hint, I suggest effective is the better target).

  Listening to customers is important but it is not sufficient.

  ...

  People could assume the numbers at Enron proved what Enron was doing was correct. But it did not prove that. Until we start to evaluate data more accurately we will continue to mistakenly see proof where it does not exist.

  continue reading: The Customer Knows Best?
• Reducing Waste - Zero Waste to Landfill

  A great improvement strategy. Determine what you are trying to do. Understand the system. Set up process measures that measure that system. Improve the system and track the results. Repeat. Maintain focus over the long term. Result: success.

  continue reading: Reducing Waste - Zero Waste to Landfill
• The Exciting Life of Industrial Statisticians

  All of this provides great new opportunities for industrial statisticians to serve as statistical leaders-a term popularized by the late and great Ed Deming (see Hahn and Hoerl, 1998). Statistical leaders engage principally in leveraging statistical concepts and thinking (see Hoerl, Hooper, Jacobs and Lucas , 1993), and focus their activities on mentoring and supporting the most business-vital and technically challenging problems dealing with getting the right data, and converting such data into actionable information.

  continue reading: The Exciting Life of Industrial Statisticians
• Distorting the System or the Data is Easier Than Improving the System

  3 ways to improve the figures: distort the data, distort the system and improve the system. Improving the system is the most difficult.

  When people mistake the data proxy for the thing to improve they focus on improvement of how the data looks not of the system. That is the wrong strategy. The correct strategy is to focus on improving the system and as a way of verifying results you then look at measures. But you must always remember those measures are not the end they are an attempt to measure the end you are trying to achieve.

  continue reading: Distorting the System or the Data is Easier Than Improving the System
• Gladwell (and Drucker) on Pensions

  The most influential management theorist of the twentieth century was Peter Drucker, who, in 1950, wrote an extraordinarily prescient article for Harper’s entitled “The Mirage of Pensions.” It ought to be reprinted for every steelworker, airline mechanic, and autoworker who is worried about his retirement. Drucker simply couldn’t see how the pension plans on the table at companies like G.M. could ever work.

  Pension plans did work well for a short period of time. But recently they (along with the attached retiree health care) are one of the big problems facing large old companies: like GM. Gladwell talks about the dependency ratio for an economy and the dependency ratio of companies. Worsening dependency ratios can cause pension plans to kill companies (if they are not funded when the obligation is incurred) – as the company is forced to pay for more and more retirees with fewer and fewer workers.

  continue reading: Gladwell (and Drucker) on Pensions
• Hiring the Right People

  The job market is an inefficient market. There are many reasons for this including relying on specification (this job requires a BS in Computer Science – no Bill Gates you don’t meet the spec) instead of understanding the system. Insisting on managing by the numbers even when the most important figures are unknown and maybe unknowable. Using HR to find the right person to work in a process they don’t understand (which reinforces the desire to focus on specifications instead of a more nuanced approach). The inflexibility of companies: so if a great person wants to work 32 hours a week – too bad we can’t hire them. And on and on.

  ...

  it seems to me that the visible waste (time and money spent on the hiring process) is seen as the only waste and the much more difficult to see waste of hiring the wrong people is ignored.

  continue reading: Hiring the Right People
• Ackoff’s F-laws: Common Sins of Management

  Managers who don’t know how to measure what they want settle for wanting what they can measure.

  continue reading: Ackoff’s F-laws: Common Sins of Management
• Customer Un-focus (due to misunderstanding data)

  Data, such as participation rates can be used as in-process measures to help you locate areas to look at for improvement. Don’t make the mistake of managing to the measure. The measure should help you manage. Improving the number is not the goal. Improving the situation that the number is a proxy for is the goal.

  continue reading: Customer Un-focus (due to misunderstanding data)
• Google’s Answer to Filling Jobs Is an Algorithm

  Google can do some things well that are unwise for others to try.

  ...

  I can’t see this as an effective strategy for most companies. And even for Google, I don’t see it as a great idea, but trying ideas that might seem crazy can be an effective innovation strategy.

  continue reading: Google’s Answer to Filling Jobs Is an Algorithm
• Transforming Management: Risks and Importance of Data and Quality Tools

  ...Often measures become the focus and the reason for improvement is obscured. Improvement should eliminate waste and improve value to the customer. Measures should help determine the success along that path but improving the numbers is not the aim, the numbers are merely proxies for that aim.

   

   

  continue reading: Transforming Management: Risks and Importance of Data and Quality Tools
• Systemic Failures Lead to Many Fatal SWAT Raids in the USA

  ...people being killed in raids by police on the wrong house: police in full swat gear storming the wrong house by accident and then killing occupants. The media in general sees these as “special causes” – isolated incidents. So while tragic the strategy is then to examine what mistake in this unique situation lead to tragedy. I believe this is a systemic problem and therefore see the proper examination to undertake is to look at the whole system. That is, to use the common cause improvement strategy – when the tragedy is seen not as an isolated incident but the result of a system.

  continue reading: Systemic Failures Lead to Many Fatal SWAT Raids in the USA
• Metrics and Software Development

  I find looking at outcome measures (to measure overall effectiveness) and process measures (for viewing specific parts of the system “big picture”) the most useful strategy.

  The reason for process measures is not to improve those results alone....

  process measures are helpful in serving as indicators that something is going wrong (or potentially going better than normal). Process measures will change quickly (good ones can be close to real time) thus facilitate immediate remedies and immediate examination of what lead to the problem to aid in avoiding that condition in the future.

  continue reading: Metrics and Software Development
• Data Visualization Example

  In Myths About the Developing World, Hans Rosling shows some great graphics to display data on health care outcomes.

  The presentation also gives a concrete example of faulty knowledge (people thinking things which are not so - related to theory of knowledg). See gapminder.org for good additional material.

  continue reading: Data Visualization Example
• Be Careful What You Measure

  Behavior can be changed by what is measured. The problem with arbitrary numerical targets (to take one measurement related example) is not that attempts to achieve those targets won’t have an affect. They very well may have an affect. However they may not lead to the desired result. When focused on improving a number (which can happen when focused on measures – especially as the focus on those measures is tied to bonuses, favorable treatment…) the focus is not necessarily on on improving the system. Often distorting the system is the result.

  continue reading: Be Careful What You Measure