Lesson 3: Paint and Pretty Pictures

Child's PlayEverybody gets so much information all day long that they lose their common sense.
~ Gertrude Stein

We have talked about sources of data and information. How to sort out the wheat from the chaff and focus on the information that is important to you.

Now you have done that and you are clear about what is going on, the next step is to be able to explain it to everybody else.  How are you going to present the data?

How do you show the information in a way that tells a compelling story?  A way that keeps your audience on the straight and narrow?

What is Your Point?

In 2009 the Local Government Association released a list of 200 words and phrases.  Words that councils should avoid as they are unclear.  Included on the list were the following terms and their recommended replacements:

  • Actioned – Do
  • Core message – Main Point
  • Functionality – Use
  • Outcome focused output – Result
  • Potentialities – Chances
  • Procure – Buy
  • Slippage – Delay
  • Transformational – Change

Leeds City Council rose to the challenge, slashing the statement:

“Individual Chief Officers have the delegated authority to appoint employees on a temporary basis to provide cover for staff absences, to cater for peaks in workload and to deal with any tasks which may arise which are outside of normal workload of the department. Such approvals to be subject to there being budgetary provision available and a check being made with the Personnel and Training Division to ascertain whether the cover could be provided by either a redeployment or a secondment.”

Down to a simple:

“If they are required, money is available, and the job can’t be done by another member of staff, chief officers may hire temporary workers.”

You might be thinking what on earth has that got to do with data presentation?

Quite simply it is about communication.  Every chart or table that you put up should give a clear message.  Have a look through your old presentations and ask yourself one question.

“What exactly is my point?”

If you can’t answer that question don’t even think of drawing a graph.

Choose Your Weapon

Once you know what the point that you are trying make is, it is time to choose the mechanism that does it best of all.

Data tables

Accountants love data tables; I think all those numbers give them a sense of security.  A good first question is should I use a table or a graph?

Tables work when:

  1. It is necessary to look up each value.
  2. The precise value is important.
  3. There are multiple units of measure.

If you are in a hospital and the doctor is looking up the dose of a drug he is about to administer;  you’d hope he is using a table of data to do it.

If the message is all about the shape of the data and you are trying to show relationships a graph is better.  A picture tells a thousand words.  Look at the information on military spending in the table and chart below.  Which is getting the message over?

If you have decided to draw a graph, the next question is which one to use.  The different software houses are in mortal combat.  They want to provide every colour, shape and size available.  But it isn’t helpful, there are only 4 types of graph you need:

Pie charts

The simplest type of chart.  They are useful for displaying portions of a whole.  E.g. how employee’s spend their time at work.  Pie charts are simple and easy to understand.

The pie chart has two drawbacks.  If there are more than about 5 or 6 different slices they become hard to read.  If the values of components are similar it is difficult to see the largest.Pie Chart

Line charts

Use a line chart to show how something varies over time — to show a trend.  Normally, time is shown along the horizontal axis.  Show the other unit along the vertical axis.  The resulting line indicates a continuous flow over time.

When using line charts be clear what the scale of the vertical axis is otherwise it can confuse.  The two lines on the chart below show exactly the same data but on different scales.  Changing the scale can give a very different message.  Is performance stable, or all over the place?

Line Chart

Bar chats

Bar charts are best used to show data from different categories.  E.g. volume of calls by call reason or number of orders by product type.  A line chart implies a relationship between points as it moves across the page.  E.g. sales volumes over time.  Bar charts show different items which are not related.  E.g. sales of different products.

Ordering items from largest to smallest is a good way to show relative importance.

Bar Chart

Scatter plots

Use scatter plots to show the relationship between two variables.  Perhaps height and weight, or price and number of sales.

If the points show a pattern then there is a relationship, if they don’t then it is safe to assume there isn’t.  Scatter plots can be useful when trying to determine cause and effect.  For example a scatter plot will show if the size of a sales order determines the time it takes to process.

It is best not to show more than one relationship on a scatter plot at a time.  If you do you will just see dots before your eyes.

Scatter Plot

Be as Dull as You Like, in Fact Be Duller

You wouldn’t have a serious conversation with Motörhead blaring away in the background.  Not unless you were 16.  The noise is distracting.  It gets in the way.

It is just as hard to hold down a conversation with a couple of screaming kids in the background.  I have tried.  The conversation gets diluted, communication doesn’t happen.

This is obvious, blindingly so.  Studies have shown that most people have an attention span of about 8 seconds.  That is 8 seconds for people to get the point that you are trying to make.  So why do we insist on adding lines, points, annotations and all sorts of other stuff to the charts we draw?  Edward Tufte has a nice turn of phrase for this; he calls it “Chart Junk”

The purpose of a chart is to communicate an idea or point.  If you doll it up with 3D and colour, axis lines and fancy graphics, you are just adding noise.  Stick to the point, less is more.  The only exception is to add axis labels and scales (just like your teacher told you to when you were 10.)



Review the management information that you have and ask yourself:

  1. What is the point that each chart is making?  Is the message clear from the moment you look at it?  If it isn’t work out what that message is, write it at the top of the chart.
  2. Look at the data you have.  Work out which of the data sets you have is critical to supporting the message.  Remove everything else.
  3. Determine which of the basic chart formats is the best (clearest) way to reinforce your point.
  4. Where a specific action or issue needs to drawn out, state it at the bottom of the page.
  5. Remember this isn’t an exercise in page optimisation, it is an exercise in communication.  It doesn’t have to look pretty or managerial, it does have to look clear.

In the next lesson we will talk about how you use management information.  It is useless if it doesn’t drive management action.

Thank you for reading.

I’m not dumb. I just have a command of thoroughly useless information. ~ Bill Waterson

Further Reading

I hope this article has demonstrated that there is an art to presenting information.  You should do it in a way that leaves your audience absolutely clear.  They must understand your message without any confusion.  To learn more, the simplest (and best) book I have found is “Say it with Charts” by Gene Zelazny”.

It “provides easy-to-follow tools and strategies for creating powerful business presentations, guiding you through what to say, why to say it, and how to say it for the most impact.”  You can find out more from the Amazon link below:


Post Script

Did you get here from a link from a friend, Facebook, or Twitter? This lesson is part of a 16-part free e-mail course. Learn more about it and sign up here.