The problem with metrics

By | July 11, 2015

Metrics are everywhere. As a consultant, we are in the business of quickly understanding business problems, proposing solutions, and measuring results. Metrics are a necessary evil of any project. The classic continuous improvement methodology DMAIC depends on M (measuring) and C (controlling the process).  Seemingly, Consulting is all about metrics and key performance indicators (KPI).

Metrics are good.  Without question, they are an invaluable way for executives to track their relative performance (against competitors, and their own historical performance), manage their business, and focus their attention. There is clearly a reason the government publishes thousands of economic indicators monthly, and why Wall Street analysts brood over companies financial and valuation statistics daily. Financial statement analysis (FSA) is one of the most popular and powerful classes that MBAs take in business school.  Variable A divided by Variable B = A/B = a metric.

Look at these two contrasting quotes from Peter Drucker, genius.

What gets measured gets managed. – Peter Drucker

 

There is nothing so useless as doing efficiently that which should not be done at all. – Peter Drucker

 

Metrics are bad. What’s wrong with them?  Using the previous logic, shouldn’t it be manna from heaven?  Aren’t more the better?

While there are many reasons why metrics are important to run a business, compare performance with competitors, and provide you trend analysis, I can also think of 10 reasons why they often make you stupid, lazy, or worse.

1. There are too many of them.  If you look at the dashboard some executives, you will see that there it is an printed out excel spreadsheet with 12 size font and lots of lines. People are inundated.  It’s too much to really comprehend.  A better approach is to really think through the problem and distill into a handful of key metrics that can cascade down to more detailed levels of measurement.  Seeing a CEO have a list of dozens of metrics is annoying and a bit sad.

Some regulated industries are the worst because there are piles and piles of metrics required by regulators, industry watchdogs, investors, board members, employees, and bankers.  Eventually, executives gets numb from the onslaught of numbers.  As I tell my clients, strategy is as much what you don’t do (and focus on), as what you do.

2. Metrics are not created equal.  Too often, metrics are given the same level of importance.  You will see a dashboard with 3-4 mission critical ones alongside something almost trivial in its minutiae. It turns into a dustbin of measurement.

If you need to see an example of this, go to healthgrades.com and look for a hospitals near you here.  Under quality you will see things like mortality and readmission rates (important) next to things like % of patients who receive aspirin at discharge (yes, it is a CMS-mandated measure).

3. Some measure the past. There are leading and lagging metrics.  Some metrics (e.g., market share) look at the past.  They are lagging the current reality, and if you are not careful, you start managing your business based on history.  We know that very few things in life are straight line, so it’s like driving your car on a curvy road looking in the rear view mirror.  Try to mix both lagging and leading metrics.

Consultantsmind Dashboard - The problem with metrics

4. Some can give a false sense of causality.  Metrics are – by their definition – reductionist. They give an indication of something. They are the symptom, not the root cause. Continuing with the driving analogy, the speedometer tells you how fast you are going, but does not tell you whether you are going to the right place or if that speed limit is appropriate for the car’s engine, or the area you are driving in.

5. Some can give a false sense of achievement. I believe executives often look at metrics too frequently.  Yes, it comes from #4 (false sense of causality), but it also gives the self-imposed illusion of doing something about the problem.

Typical setup:  An executive spends hours looking at national market share metrics; she calls meetings to review, manage, align, discuss, debate the market share numbers. It’s fairly typical behavior, and yet, there is a weak chance that the activity of the last 29 days (since the last time you checked) is actually affecting market share nationally. Lots of discussion on the speedometer is not making the care go any faster.

6. Some can use the wrong benchmarks.  It is very easy to compare your performance against the wrong peer set.  Like a boy who tells his mother that his “B” in math was not bad because all his friends got “C”s, who you benchmark your performance against matters. Simply put, you cannot compare the operational performance of companies or business units that are in different industries, stages of development, or have capital structures. Don’t compare apples and oranges.

7. Metrics don’t give advice.  Metrics just tell you the situation, they don’t tell you what to change.  It’s like smoke detector that is screaming at you, but honestly, it is not telling you how to escape the fire – that is what you have to do.  Too many people waste their energy on lagging indicators, and don’t spend enough time thinking looking at leading indicators of where customers are going and how to fight differently.  

8. Metrics can be a game.  Incentives are often based on metrics because it is a concrete way to keep score.  As a result, smart people game the incentive system to their benefit – pushing out/pulling in bookings and billings to suit the compensation plan. In business slang, they call it “sand-bagging” and it happens all the time.

9. Some waste time.  It is incredible how much time is spent collecting data to put into a report to track metrics.  Like bad legislation, reports in the corporate world rarely die, instead, people spend time calculating metrics which may / may not be used in decision making.  If you talk to line-staff at a company, they often complain about the onus of tracking things which have no meaning.  The office version of Sisyphus.

What to do? As a consultant, you should give your clients good advice on metrics. Help them to think about the problem.  Be wise and don’t fall prey to the metric-mania which runs many companies and offices:

  • Have a point of view on which metrics a client should focus on.
  • Balance lagging and leading indicators
  • Question why some are being used? Test the assumptions
  • Create metric hierarchies so that they can roll up to the CEO, and down to each functional department.  Give it some systemic design
  • Bring the client back to basics and ask them more strategic questions
    • What are they trying to achieve?
    • What does success look like?
    • What is the relative time frame to make things happen?
    • What are some leading indicators to look at?
  • Don’t fall into the trap of SG&A review of unhelpful ones

Related posts:

4 thoughts on “The problem with metrics

  1. Sandeep

    Good advice man…ironically, I am helping a CEO think through metrics for his organization as we speak…very timely reminders for me 🙂

Comments are closed.