Working In Uncertainty
Visualizing an uncertain sales pipeline at Z/Yen
by Matthew Leitch 21 May 2010, with help from Z/Yen
Z/Yen is a small but influential consultancy based in the City of London with a track record of coming up with innovative ideas for just about anything, so it is no surprise that their internal management systems are full of interesting ideas.
Their approach to visualising their sales pipeline is typical of them but not typical. Like most consultancies, Z/Yen's income often arrives in large lumps, whose arrival, timing, and size are highly uncertain. Z/Yen's directors, Ian Harris and Professor Michael Mainelli, wanted a method that would give more information than just a list of sales in progress with best guesses for their value.
Their system produces various displays that show the probability distribution of sales out across future periods. (In fact it is a sales tracking system with a lot of other functionality, but it is the visualization of sales forecasts that is the subject of this Innovative Case.)
The system has been used monthly by Z/Yen for several years. Internally it is known as Z/EAL. It has also been implemented by a number of clients and is marketed under the names PPRISM and PPRIFM.
In this first picture you can see the total sales actually agreed each month (red blobs), their annual moving average (the fairly steady line with black squares), and various predictions of expected value and high and low values month by month. In this example the sales forecasts seem to have been somewhat optimistic, as revealed by the weighted 12 month rolling average win value, which is typically lower than the weighted expectation.
This second picture shows another style of display. Here the coloured bands indicate confidence intervals, showing in effect a sequence of probability distributions over time. In this example the uncertainty in forecasts is fairly low – much lower than in the first picture.
The tool is built with Excel and Access. It calculates the distributions by applying Monte Carlo simulation to the database of possible sales, using the probability distribution information for each sale, which is captured using triangular distributions. In the example below the system has run just 1,000 scenarios, but this is enough to show that a wide range of outcomes is possible.
The tool can provide non-probabilistic displays too. The picture below cuts potential sales into different types of work and different stages in the sales pipeline.
Why this is a good idea
The information in these displays helps Z/Yen make decisions about things like when to get associated consultants involved (because their own employees may be unavailable), when to take holidays, when to stop selling, and when to sell harder. They can also learn about their forecasting tendencies.
All this is on the basis of displays that counter the natural tendency towards uncertainty suppression (i.e. acting as if more certain than you are or should be). This helps Z/Yen stay aware of the value of flexibility and of when to get more information to help with difficult staffing and pricing decisions. Without this perspective they would feel constantly surprised by the difference between forecasts and actual sales.
Z/EAL is also a classic example of an intelligent internal control. It involves explicit thinking about uncertainty, modelling, and management decision making. It leads to additional risks responses being put in place, usually in the form of information seeking and flexible agreements.
Finally, it is an outstanding example of fully embedded risk reporting. Rather than reporting risks on a report separate from sales forecasts, Z/EAL simply shows the uncertainty within the forecast report itself.
Further informationFurther information is provided by these articles on the Z/Yen website.
I would like to thank Z/Yen for permission to feature their innovative system on my website and to thank Michael Mainelli for introducing me to Z/EAL and providing information on its use.
Hundreds of people receive notification of new publications every month. They include company directors, heads of finance, of internal audit, of risk management, and of internal control, professors, and other influential authors and researchers.
Made in England
Company: The Ridgeway Expertise Company Ltd, registered in England, no. 04931400.
Registered office: 29 Ridgeway, KT19 8LD, United Kingdom.
Words © 2010 Matthew Leitch