Let’s assume you have a target to increase sales of ice creams by 10% over last year’s sales, by value. Each month you receive a report which shows the latest sales:
Sales for the last six months
This is accompanied by a longer report which splits each of the categories into individual products such as King Cones, etc.
What does this report show? Well you’re in trouble – half way through the year and sales of ice creams have actually decreased. Can you base any decisions on this report, and the more detailed report which accompanies it?. No, not without asking for more information.
Why, in our example, do you require more information? Because you don’t know why sales are low, what decisions have already been taken and what decisions are expected from you.
So what information should you have been given? Well, for a start, you don’t need to know sales of chocolate bars and toffees because you are not responsible for them. You need more details, so you go to the longer report. This shows the sales of 300 products and, by closely reading it you notice all have sales increases of 5% except the three best sellers which show sales decreases of 20%. But you still can’t make a decision because you don’t know why sales have decreased. You ask your staff, who tell you that production problems resulted in no deliveries to shops in the last month and your decision is required whether to demand compensation from your supplier. However, the report gives none of this information.
What’s the alternative? You only need to know the sales of products where they fall outside the targets you set (e.g. if sales increase by less than 5%). Where they do, you want an explanation from your staff, details of decisions they have taken and, possibly, what decisions they want from you.
Wait a minute; won’t that result in longer reports? Well you’ve removed all the information not requiring a decision and you could receive a verbal, not written, report.
This approach means that your staff, who probably prepare the report on a spreadsheet, also take more interest in its production, since they have to explain differences. It also gives them the opportunity to ask for a decision, which staff are sometimes reluctant to do.
Your staff probably work overtime to produce this report by very tight deadlines. You probably look at it when it arrives, and maybe ask questions to find out reasons for the poor sales, but it is very unlikely that any decision is taken as a result of the information for many days. If the decision is to demand compensation, it hardly requires staff to work overtime in order to give you the report a day early.
Let’s suppose you now have to provide forecast sales and profits, after all expenses, for the next three years. Well, you assume sales will be £500,000 this year; profits will be £20,000 with a year on year increase of 9%. Put the figures into a spreadsheet and you can go home happy:
Except that, while the figures for year 2014 have a realistic accuracy, the figures for 2015 and 2016 are quoted to an accuracy far greater than possible. However, because they look accurate they give a false sense of security, whereas more realistic figures would give the impression of guesses. Which is, of course, precisely what they are.
Your boss asks you if it’s worth carrying out an advertising campaign, which means that the profit increase has to cover the cost of the campaign. You could start trying to estimate what the sales increases from various types of campaign (TV, radio, newspapers) would be and calculate profit from these to give an idea how much you can spend.
Try working the other way round. You can’t determine sales but you can set the cost of the advertising campaign. Thus you know this figure accurately. So using data from previous campaigns work out the best and worst cases for sales and profit increases for advertising at different costs. Draw a graph, which shows that you get the greatest benefit (i.e. profit increase as a result of the campaign, less costs) for a campaign costing £40,000. At best you will get an £8000 benefit, at worst £200.
