As we have seen, data is only really useful to you if it is in context and it is analysis that adds that valuable ingredient. Now, we’re going to look at the process of analysis, the activity that converts the numbers on your dashboard into meaningful information. This information can be in the form of graphs, diagrams, illustrations, whatever you find most useful to fuel your thinking process (see Chapter 3: Planning Properly), producing the plans you need to steer your business.
Collecting and analysing the data generated allows you to demonstrate to yourself and your team that you are surviving; it provides a foundation to your plans and the instructions you issue, and prevents your pronouncements from being dismissed as a ‘load of hot air’! In addition, the analysis of the data allows you to identify opportunities for improving your business (making things better), which we’ll deal with in section 6.3.
As you are in charge, it falls to you to decide how you want the analysis of which data to be done. Obviously, you’ll discuss this with your team (they’re the ones with the specific expertise, after all), but your focus must be on what you want the analysis to tell you? If you’d like tighter control of the process, then you can write down the process in the form of a standard operating procedure. It takes effort to do this, but the benefits are that no matter who is doing the analysis, they will do it in the way that you want it done every time.
In addition, you can get your people to record their measurements on the dashboard and show the working, then it can be checked. This is a good way to improve reliability as people who know that others will be looking at their work tend to take more care over it. Depending on your business, you can use a specially created form on paper or one in an electronic spreadsheet. There are many spreadsheet programmes available and they come with built—in calculators and equations, so that once you’ve created the spreadsheet, anyone can fill it in and it’ll provide a consistence and, most importantly, accurate answer.
In Chapter 5: Setting Up Your Dashboard, we looked at some of the most common business measurements and why they might be important to you. What we’re going to look at here is an example of what can go wrong, if you stop analysing your data. Table 6.3 shows the sales data (in £s) and the number of customer complaints received at Generic Boxes Ltd (GBL) from the beginning of their first year.
| MONTH | SALES (£) | COMPLAINTS |
| January | 10,000 | 1 |
| February | 15,000 | 2 |
| March | 20,000 | 3 |
| April | ||
| May | ||
| June |
Table 6.3: Incomplete first half—year results for Generic Boxes Ltd
Let’s do some analysis (in this case, merely having a good look at the data) and see what, if any, information can be extracted. Well, first of all, it appears that their sales are growing at a consistent £5,000 per month, which is fine and provides a good base from which to build a thriving business. However, the monthly complaints are also increasing and this could present a very real threat to GBL’s survival. Now, we don’t have any additional context, but we can suggest explanations for the data, which can be accepted, refined or discarded when further information is obtained:
- As the sales rose, GBL didn’t take on extra personnel to cope with the increase in workload. This led to increasing delays, which is the source of the increase in customer complaints.
- As the sales rose, GBL did take on extra personnel, but there hasn’t yet been enough time to train them to competence properly. This has led to production delays and the subsequent increase in customer complaints.
- As the sales rose, GBL lost personnel due to stressful working conditions and haven’t yet had time to replace them, which is the source of the increase in customer complaints.
You can see that analysis of the data produces information that can be used to make decisions regarding how to proceed, but there are a number of different ways to interpret the data, so leaping to conclusions could be unwise. Maybe, GBL should gather more data to help with the analysis and interpretation? This would be a wise response to a perfectly normal situation for a start—up business to be in.
Be aware that the amount of data and context available to an early stage business is limited and so the confidence levels with which decisions are made is lower, relatively speaking, when compared to more established organisations. The absence of more data, more context and more information means that the managers at GBL are, to a greater or lesser degree, ‘flying by the seat of their pants’. However, as their business grows, more data will be available, greater context will be present, more information can be extracted from the data, and their decision—making will become more confident. They just have to survive long enough.
Okay, so let’s move forward in time and see how Generic Boxes Ltd are doing by the end of their first half—year.
| MONTH | SALES (£) | COMPLAINTS |
| January | 10,000 | 1 |
| February | 15,000 | 2 |
| March | 20,000 | 3 |
| April | 30,000 | |
| May | 20,000 | |
| June | 10,000 |
Table 6.4: Complete first half—year sales results for Generic Boxes Ltd (incomplete complaints data)
The observant amongst you will have no doubt noticed that Table 6.4 is supposed to be complete, but it isn’t. There has been a problem. The small team at GBL got very busy very fast. As a result, time was tight and, whilst they continued to collect some data (their sales figures), they didn’t bother to keep measuring the complaints they received. They reasoned that one or two complaints a month wasn’t something to worry about and they were very busy, so all must be going well. They let their analysis slide, they took their eye off the ball, and they ceased to bother continuing to measure the complaints. Could this be a mistake?
At the end of six—months, the pressure has eased and they suddenly have the time to sit down, meet formally and do some proper analysis. They are no longer ‘too busy’ as their sales have declined and they are about to find out why! They’ve gone back over their records and filled—in the half—yearly complaints results. Can we find any explanations for the state of GBL’s business?
The complaints data shows that there has been a tenfold increase in the level of monthly complaints since the beginning of the year. This could indeed explain the downturn in their sales figures. Let’s do some off—the—cuff analysis, provide possible interpretations and look at what GBL could have done to avoid this potential disaster.
| MONTH | SALES (£) | COMPLAINTS |
| January | 10,000 | 1 |
| February | 15,000 | 2 |
| March | 20,000 | 4 |
| April | 30,000 | 5 |
| May | 20,000 | 6 |
| June | 10,000 | 10 |
Table 6.5: Complete first half—year sales and complaints data for Generic Boxes Ltd
First off, let’s assume that all the complaints came from the same customer. This means that they upset one of their customers repeatedly and increasingly in the space of six months. Now, this situation looks bad for GBL if they have only a few customers or the one customer they have upset is (was) their biggest customer in terms of sales. They may be able to rectify the situation, but they’d better move fast.
Alternatively, we could assume that all the complaints came from different customers. This means that they have upset 27 of their customers in the space of six months. Again, if they have many hundreds or thousands of customers, then this complaint rate might not be too bad, but just look at the sales figures by June; they are going down as the complaints are going up. Do you think that there might be a link?
The managers at GBL looked at their early sales figures and got excited. Unfortunately, the excitement that they felt at their increasing sales made them believe that ‘one or two’ complaints a month wasn’t worth bothering about. As a result of making poorly—informed decisions, they made their first customers unhappy. These customers promptly took to social media to drag GBL’s name through the mud. The business had only been in existence for six month and already it’s looking doomed to failure.
To be fair, looking at increasing sales figures and getting excited is to be expected, we all do it. However, you must never forget that sales are the result of customers buying from you and then, usually, coming back to buy from you some more, unless you sell to each customer only once. Repeat business is one (but only one) of the measures of the loyalty that you have established with your customers. However, it is often how you deal with complaints that can differentiate you from your competition.
Your customers may decide that the way that you deal with them when they are unhappy is a far better measure of your worth as a business than how you take their money off them when they are happy. Give them an excellent service when you have upset them and you are more likely to have them “singing your praises” because they understand that ‘when the chips were down’ you cared.
It is beyond this scope of this course to teach you how to do statistical analysis, but there are some simple ways to collate and analyse your data that are applicable and useful to just about any business. I’m talking here about averages and percentages. Now don’t worry, I’m not going to start explaining in detail the statistical theories behind how you calculate averages or percentages, but what I do want to address here are some of the associated pitfalls if you use statistics incorrectly. We’ll start with averages.
Like all management tools, when deployed correctly, knowing the average of a set of numbers can be very useful to you. What is an average? Well, the average of a set of numbers is derived by totalling all the numbers in the set and dividing that total by the number of individual numbers in the set. So, why should you be cautious when using averages in your decision making processes? Well, let’s use a set of hypothetical unit sales figures for a firm of accountants, Additup Accountancy Ltd, as an example.
| JAN | FEB | MAR | APR | MAY | JUN | TOTAL | AVERAGE |
| 150 | 142 | 196 | 131 | 215 | 166 | 1,000 | 166.67 |
Table 6.6: Monthly unit sales figures for half a year for Additup Accountancy Ltd
In this example, let’s say that each unit sale is equivalent to Additup Accountancy Ltd preparing one set of accounts, so in January they prepared 150 sets of accounts, in February 142 sets, and so on. The two columns on the right hand side of the table show the total unit sales and the monthly average. There is nothing particularly remarkable about this set of numbers, other than the fact that they add up to 1, 000, which is for convenience in this example. Now, that all seems very obvious, so where’s the pitfall? Well, we’ll add a few extra rows containing the half—yearly sales figures for a number of other accountancy firms to illustrate the point.
| JAN | FEB | MAR | APR | MAY | JUN | TOTAL | AVERAGE | |
| Additup Accountancy Ltd | 150 | 142 | 196 | 131 | 215 | 166 | 1,000 | 166.67 |
| Ashworth and Co | 166 | 167 | 167 | 167 | 167 | 166 | 1,000 | 166.67 |
| Bouncing Checks LLP | 1000 | 0 | 0 | 0 | 0 | 0 | 1,000 | 166.67 |
| Massive Accounts | 500 | 0 | 0 | 0 | 0 | 500 | 1,000 | 166.67 |
| Old Mr Knibs Ltd | 50 | 150 | 300 | 300 | 150 | 50 | 1,000 | 166.67 |
| Zebrafish Ltd | 300 | 150 | 50 | 50 | 150 | 300 | 1,000 | 166.67 |
Table 6.7: Monthly unit sales figures for half a year for multiple businesses
A quick analysis of the sales figures in Table 6.7 tells us the following:
- Additup Accountancy Ltd — sales go up or down each month through the half year
- Ashworth and Co — sales are more or less constant month—on—month
- Bouncing Checks LLP — all sales occur in January with nothing else through to June
- Massive Accounts — half the sales are made in January and half in June with nothing in between
- Old Mr Knibs Ltd — sales increase from January to March and then decrease to June
- Zebrafish Ltd — sales decrease from January to March and then increase to June
Notice anything? Yes, the sales figures across the rows are all different and yet the totals and the averages are all identical. If you were to use the averages alone, then it would not give you any information that you could use to differentiate these businesses. However, the interpretation of the data in the total and average columns is helped enormously by using the information provided by looking at the monthly sales figures in detail. Taken alone, this is why there are pitfalls to using averages.
Those of you with a good grasp of statistics will be aware of other measures, such as standard deviations, standard errors, medians, modes, etc., that you could use but, as I said, those details are beyond the scope of this course. Suffice it to say that you must be careful when analysing your data and, as a general rule, the more data that you have, the better.
The other value that appears frequently in business that I want to address is the percentage (%). The pitfall here concerns the difference between the percentage (%) and percentage points. As with averages, percentages (%) and percentage points are management tools and providing that you deploy them with understanding and a dose of common sense, they’ll help you to make informed decisions.
So, what’s the difference between a percentage (%) and percentage points? Well, a percentage (%) is always determined in relation to the total, literally ‘per hundred’, where the ‘hundred’ is taken to represent the whole of whatever it is that you are measuring. Percentage points are the individual percentage units themselves. To illustrate, let’s go back to Generic Boxes Ltd’s (GBL) and look at their first quarter sales figures.
| MONTH | TOTAL SALES
(Units) |
MARKET
SECTOR A |
MARKET
SECTOR B |
MARKET
SECTOR C |
| January | 1,000 | 500 | 10 | 490 |
| February | 2,000 | 1,500 | 40 | 460 |
| March | 3,000 | 2,400 | 80 | 520 |
Table 6.8: Generic Boxes Ltd monthly unit sales by market sector
A little analysis of Table 6.8 reveals that GBL’s total sales increased between January and February from 1,000 to 2,000 units. Therefore, their sales doubled, which is an increase of 100% from January to February. Further, we can see that their total sales also increased between February and March from 2,000 to 3,000 units; an increase of 50% from February to March. Note that the respective percentage increases are in relation to the previous month’s sales.
Now, let’s have a look at the market sectors. In January, 50% of GBL’s total sales came from market sector A (500 out of 1,000), in February it was 75% (1,500 out of 2,000) and in March it was 80% (2,400 out of 3,000). So, between January and February, the contribution of sector A went from 50% to 75% of total sales, an increase of 25 percentage points. When compared to sector A’s total sales contribution of 50% in January, this 25 percentage point increase in February represents a 50% increase in contribution to total sales, 25 being 50% (or half) of 50. Similarly, between February and March, the contribution of sector A went from 75% to 80% of total sales, an increase of 5 percentage points. When compared to sector A’s total sales contribution of 75% in February, this 5 percentage point increase in March represents a 6.67% increase in contribution to total sales, 5 being 6.67% of 75. You get the idea
We’ll use market sector B’s result as our final example. In January, 1.00% of GBL’s total sales came from market sector B (10 out of 1,000), in February it was 2.00% (40 out of 2,000) and in March it was 2.67% (80 out of 3,000). Between January and February, the contribution of sector B went from 1.00% to 2.00% of total sales, an increase of 1 percentage point. When compared to sector B’s total sales contribution of 1.00% in January, this 1 percentage point increase in February represents a 100% increase in contribution to total sales, 1 being 100% of (or equal to) 1. Similarly, between February and March, the contribution of sector B went from 2.00% to 2.67% of total sales, an increase of 0.67 percentage points. When compared to sector B’s total sales contribution of 2.00% in February, this 0.67 percentage point increase in March represents a 33.33% increase in contribution to total sales, 0.67 being a 33.33% (a third) of 2.00
And there is the pitfall. Knowing that you’ve increased your sales by 100% on the previous month is great news, news to get excited about. Until you realise that it only equates to an increase of one (1) percentage point! You may have read in certain sections of the media that eating or drinking this or that will increase or decrease your chance of developing a terminal illness. To make up a daft example, let’s say that the article states that drinking a glass of lettuce juice every day will reduce your probability of growing another head by 50%. When you look at the background information you discover that the probability of growing another head in the general population is, say, 2 in 100,000. Drinking the lettuce juice daily reduces this to 1 in 100,000. Sure enough, reducing 2 to 1 is halving it and a half is 50%, but as a percentage of the 100,000 the results are going from 0.002% to 0.001%, a change of just 0.001 percentage points! Let this be a lesson to you. Be aware of spurious claims and be careful when analysing your own data.
Lies, damn lies and statistics
If your analysis is to produce the information required to steer your business through survival to success, then the data that you use must be in the correct context. If you take any old set of numbers in any order, then your information will be poor, and so will your thinking. In addition, whilst a quick glance at your dashboard every now and again is better than nothing (usually), regular, planned reviews are better.
Analysis allows you to get at the vital information sitting in your business. Without this information you will be missing much that can help you with your thinking and the steering of your business. Having measured your achievements and analysed your data, the information generated must be made available to those who can use it to your business’s best advantage, and we’ll look at how to do that in the next section.