Thursday, April 30, 2009

Why benchmarking is dumb

Benchmarking is the process of assessing the abilities of your company/team/department against a similar organization that competes with your own. It is not only dumb it can be highly dangerous as well. Here are three reasons why.

Your competitive advantage comes from what you do differently to your competition. As Tom Peters famously said benchmarking is the process by which you make a 3 year plan to catch up to where your competitor is today. It is dumb because it focuses you on those things that are similar to your competitor instead of what is different. It is dangerous because it leads you to believe that you're going in the right direction when in fact you're going backwards.

The second reason benchmarking is dumb is that it stifles ambition. It dumbs us down into believing that if only we were just a little better than our competitors, then everything will be OK. So we work hard on understanding what our customers want and then deliver it just a little better than our competitors. This is dangerous because it leads us to believe we have a sustainable competitive advantage, when in fact all our competitors need to do is to run a little faster to overtake us again. This kind of thinking can lead to a features cold war: the competitor adds a feature so we add one too just to keep up. Before you know it both have huge arsenals of features. Meanwhile someone else had a better, simpler idea and we didn't see it coming. Remember Seth Godin's words: "You cannot be remarkable by following someone else who is remarkable".

My final reason comes from what I have learned from the world of agile development. Retrospectives lie at the heart of agile methods such as scrum. The underlying principle is one of continuous improvement: inspecting what went well and what went less well, then adapting the next steps based on the lessons learned. It follows that there is no fixed agile process: I can tell you how I did an agile project but I cannot tell you how to do yours. It is the same with benchmarking. It is dumb to think that something that someone else did can be applied like a formula to your own world. It is dangerous because it can stop you thinking, stop you inspecting what you do and adapting to do it better.

Don't misunderstand me, learning from others how they do things can be hugely valuable. In particular, I think there is much to learn from people who are not your competitors but do something similar in a related world - but that's a topic for another time. The secret is not to let it stop you thinking for yourself. Inspect what others do then adapt what you do as a result of what you have learned.

PS (April 30th, 2009) Bryce Harrison alerted me to this great article on Systems Thinking and the Case Against Benchmarking. Worth reading to the end where you will find these words from Toyota "They (managers) tend to get distracted by easy-to-measure or impossible-to-emulate differences, when the really important differences lie in the harder-to-see ways value-creating activities are organised.” Their advice is clear: “To Hell with your competitors; compete against perfection by identifying all activities that are muda (wasteful) and eliminating them." Wow!


  1. Actually, your last paragraph refutes everything you said in your initial argument. And explains why you must benchmark where it is possible; to gauge not only what your customers expect in your industry, but what world class service does in others. Especially in today's specialized world, chances are that some company is providing world class service performing only the duties the department you are attempting to benchmark.

    The secret is what you do with the information. Seth is putting up a straw man. The key is not emulation or catch-up, although that is informative as well. You'd be surprised how many companies are providing sub-standard service in legal, or admin, or sales to their customers, and don't even know it. The secret is to realize where you are in relation to others, and think of ways to supersede them. But you can't do that if you don't know if you're in 1st or last place given your prices and cost structure.

    Your PS example only explains how you can do benchmarking stupidly, which we hope no one would consider. Lastly, I question having an audit group perform bench marking for a lot of reasons. And the check-plan-do exercise is a helpful one to review every few years if all it does is inspire innovation.

  2. The danger with benchmarking it is all too easy to do it stupidly. The whole concept lends itself to poor interpretation.

    Put it another way I would much rather spend time, effort and money on understanding my customers and users needs, than on comparing myself to competitors.