twitterfacebooklinkedin

Do Your Performance Reports Stack Up?

by Stacey Barr

If your performance reports are stacking up, unread and unused, then they're obviously not "stacking up" well.

– Stacey Barr

Introduction

It's an emotional thing, performance reporting. Executives give up the precious little time they have for their families to instead paw through piles of strategic reports often more than an inch thick (or they leave the pile of reports on their desk and make decisions from their guts instead). Managers earnestly trawl through operational reports (particularly looking for the measures that give their performance contracts teeth) to check if anything needs a bit of positive light thrown on it. Supervisors and teams cynically scoff about the volumes of time and effort they waste reporting tables of statistics that track their daily activities to audiences they never see or hear from. What makes performance reporting such an ordeal?

Problem #1: Too Many Tables of Statistics

While tables are the traditional way to present performance information, mostly because financial reports like the profit and loss statement have always been tabular, the problem is that tables help you miss important patterns or trends. They do this because they usually only show you current performance compared to last month's, and that kind of comparison is absolutely incapable of telling you the full story of where performance is headed. Tables also require you to read and hold a lot of numbers in your head, rather than quickly showing you the key patterns through visual methods like graphs. Tabular performance data takes a long time to read and interpret, and often different people can draw different conclusions from it. Also, tables encourage you to focus on details of specific individual results, not on the bigger picture of relationships (such as cause-effect or lead-lag) among a collection of results.

Solution: Use graphs that are designed to make the messages in your data, the answers to your questions, visually obvious

Graphs can take up less space than tables, they are far more interesting to look at, they convey the same (and more) information many times faster, and you have more control over the way people will interpret the data. Try using line charts to present your historic data when you are interested in understanding patterns of change (such as trends and sudden shifts) in performance over time. Use Pareto charts when you want to examine the relative impact or size of a collection of reasons or causes behind a particular performance result. Use bar charts when you only have a handful of data points in time series. Use dot charts when you want to present survey ratings. Don't use pie charts.

Problem #2: Thrown Together in an Ad Hoc Structure

If your reports have no obvious structure, such as clear and obvious section headings, then successfully fast and easy navigation through them is totally proportional to the number of times a user has looked through the report - how familiar they are with its contents.

It's easy to not notice a lack in the completeness of information in a report when it's got no logical or deliberately chosen structure, too. You just take what's in it for granted, and don't necessarily reflect on whether that is the right information or not – or if it's presented in the right order.

Without the structure having a direct link to strategic, tactical or operational objectives, it also makes it difficult to link in the information to the decision process, so you'll often see the report stacked neatly on the board room table next to each decision maker, but rarely will it be opened up at the most appropriate time to use it.

Solution: Understand the decision process the report should inform, and structure the report based on what the decisions are about

There are many ways to structure a performance report, but it should be a conscious decision, not a creative decision. For example, if your report is for monitoring progress against a strategic or business or operational plan, then structure the report around the objectives or goals or result areas of the plan. If your report is monitoring progress of a program of projects, then structure the report around the types of projects (such as maintenance versus capital, innovation versus small step improvement) or around the functional teams that own the projects (such as IT, Marketing, Sales, Research & Development).

Problem #3: A Dull, Monotonous Visual Layout

Reports that are put together with two font styles - one for the normal text and one for the headings - aren't just boring and monotonous to the eye. They are also hard to navigate, and make it quite a challenge to find the information you need rapidly and frustration-free.

In a day and age where word processing and spreadsheet software are easy to use and have a plethora of fabulous formatting tools (like colour, type faces, font sizes, borders, shading and so on), there is really no excuse for not adding a whole new layer of information to performance reports. This layer of information could be called "visual prioritisation", where the most important results are highlighted strongest to make it easy for the user to prioritise what they read first, and how to interpret it.

A dull layout and formatting design also deadens the interest of users, compels them to put it to the bottom of their reading pile, makes them feel uncomfortable or awkward using the report.

Solution: Learn the basics of good visual design so that you can make good use of white-space, alignment, fonts, colours, symbols, logos, borders, headings, titles and so on

Professional visual design achieves a few things. Firstly, it gives your report a personality or a mood, that can really encourage people to want to read it and to feel confident to read it. Secondly, professional visual design can make navigation and orientation through report many times easier, through the use of alignment and white-space and formatted headings to visually separate different chunks or types of information. Thirdly, clever formatting (using colour, type-faces and symbols) can make the most important information in the report almost jump off the page. The time to read and interpret the report ends up being a fraction of what it used to be!

Problem #4: Too Much Useless Bulk

It's getting close to becoming a natural law of performance management that performance reports only know how to grow, not how to condense. Despite the fact that we all know large reports just won't get read, we still struggle to take information out of our reports. There is story after story out there about how 3 years ago Manager Jo wanted a particular measure, say revenue earned per employee, and it's been reported ever since (even though Manager Jo only wanted it for a couple of months and has left the organisation 2 years ago anyway). After several Manager Jo's with their ad hoc measures, reports become bloated and unusable. This bulk makes navigation through the report a nightmare, it becomes an unbearable effort to produce, and the amount of data that has to be collated and analysed keeps analysts punching away at their computers until 9 p.m. for several nights after the end of month. And of course the users of the report don't really know which information they should give priority to, and which is just noise drowning out the important business performance signals.

Solution: Make a consensus decision about the types of information the report must contain, and set up guidelines for producing that information

Performance report writing tends to be a fairly evolutionary process, where it can shift and change and adjust almost every issue. Often many people that contribute content to the report, such as graphs or commentary or project progress updates, don't really know how the information will be used (most believe it won't be at all!), and therefore there is a lot of variation in the amounts and emphases of what ends up in the report. Formalise the design of the content of your report by defining each type of information that will be included (for example, graph, interpretation, cause analysis, project progress). Consider even going to the extent of writing, for each type of information, tips and examples for how to produce and present that information (for example, word count limits, specific questions to answer, format styles for graphs).

Problem #5: Using Default Excel Charts, or Whatever "Looks Good"

Microsoft Excel is certainly an easy enough graphing tool to use. While it can whip up all kind of creative and colourful graphs, it can't decide for you what is the best graph type for your data because it doesn't know or understand the question that the graph should be answering. So many performance reports that are filled with Excel default graphs (or someone's idea of a good looking graph) all too often lead to invalid interpretation of the information. Invalid interpretation means bad decisions.

All that distracting "chart junk" (like grid lines, fancy symbols, gaudy colours, three-dimensional bars and cones and linear trends lines that explain less than 10% of the trend in the data) almost turn decision making into a circus. Users experience disconnects as they work through the report, have to try to interpret so many different and complex graphs - even for the same basic type of measure (such as a trend over time measure).

Solution: Create and consistently use your own chart templates for the different types of measures and diagnostic analyses in your reports

The basic principles of excellent graph design are to keep your graphs simple, focused on a single question, choose the appropriate chart type for the data and question you have (for example, line charts for time series data, Pareto charts for cause analysis, scatter plots for correlation analysis, dot charts for survey ratings), maintain the integrity of the data you are graphing (for example, don't make differences or trends look bigger than they are by chopping off the bottom part of the vertical axis), and finally, use visual techniques that ensure the patterns in the data speak loudest (for example, don't use kooky pattern fills that make you cross-eyed, don't oversize the data points, use colours that are sympathetic to the colour scheme of your main report).

Further Reading

  • Anders Wallgren, Britt Wallgren, Rolf Persson, Ulf Jorner, Jan-Aage Haaland, Graphing Statistics and Data, SAGE Publications, 1996
  • Edward Tufte, The Visual Display of Quantitative Information, Graphics Press, 1983
  • Robin Williams, The Non-Designer's Design Book, Peachpit Press, 1994
  • Stacey Barr, Making Graphs Useful, Stacey Barr, 2004

© Copyright Stacey Barr

About the Author

Stacey Barr is a specialist in organisational performance measurement, helping people get the kind of data and information that tells them how well their business is performing, and how to make it perform better. Sign up for Stacey's free 'mezhermnt Handy Hints' ezine at www.staceybarr.com to receive your complimentary copy of her e-book 202 Tips for Performance Measurement.

We Recommend
Succession Planner

Find out how our leadership Succession Planner can assist you in managing your organization's succession planning efforts. The guide and software tool helps you initially identify your succession positions through to analyzing and reporting succession progress. Visit the Succession Planner information portal to download the free trial version and start using this powerful succession planning tool today.

Share This

  • twitter
  • facebook
  • linkedin
  • googleplus
  • gmail
  • delicious
  • reddit
  • digg
  • newsvine
  • posterous
  • friendfeed
  • googlebookmarks
  • yahoobookmarks
  • yahoobuzz
  • orkut
  • stumbleupon
  • diigo
  • mixx
  • technorati
  • netvibes
  • myspace
  • slashdot
  • blogger
  • tumblr
  • email
Short URL:http://bit.ly/xjc5M8

SUBSCRIBE NOW




Privacy