Briefings 68 and 69 showed the dangers of decision-rules based on simple ratios or on overall performance outcomes. For the consumer brand’s marketing-spend decision from recent Briefings, for example, either “Spend x% of revenue on marketing”, or “If profits fall, spend less/more on marketing”.
So what do we need instead? Well, the basic problem is that neither of these approaches focuses on the direct consequences of the decision – winning customers! So a decision on marketing spend should surely respond to the consumer win rate itself. This illustrates a useful principle:
Since resource flows determine performance improvements, decisions should take strong account of their effects on those flows.
This principle can rarely be used on its own, because spending also has a direct and immediate impact on current profits. Nevertheless, it should certainly be part of policy whenever decisions affect resource flows.
New customers may of course be won for other reasons, such as product visibility or word-of-mouth, so it is important to understand how each possible factor drives any particular flow-rate. But if we can do this, we get a further benefit of linking decisions to the resource flows they control – there is no need to wait for their effects to work all through the system, as we had to do when profit drove the decision. As soon as we know if more marketing increased the consumer win rate, we can decide whether to spend more or less next period. Greater confidence in the policy, also means we can move the decision by greater amounts. Figure 1 shows the result of starting with spend of $0.5 million per month, and raising (lowering) that spend by $100 000 each month if it resulted in an increase (decrease) of the consumer win rate.
Figure 1: Results of a policy to change advertizing spend based on change in the consumer win rate. (Click image to view larger)
The new policy has worked well. The number of consumers interested in the product has risen quickly to near its maximum level, given the tendency of some consumers to lose interest each month. This occurred because the advertizing rose relatively fast to a rate that ensured a net consumer win rate that was positive, albeit rather low in later periods, precisely because it had been so successful earlier on.
We can make further adjustments to the rule. For example, we would not want to spend another $100 000, just to win one more consumer, so we could require that the net consumer win rate must exceed some threshold of affordability.
The policy is not completely fool-proof. For example, if the spending rate is rather low, and customers lose interest quickly, an increase in spending could appear to have too little impact to be worthwhile. Management may then feel “Well, we tried more advertizing and it didn’t work, so let’s cut back again,” simply because they spent nowhere near the rate needed to win potential customers faster than others were being lost. The existence of customer segments also complicates the picture. Advertizing may appear to win new customers rapidly, but this could reflect the existence of a particularly responsive but small segment. Further increases in spending could then have disappointing results as the less responsive segment increasingly dominates the remaining market potential.
Nevertheless, the rule is much more powerful than either simplistic ratio-based rules or profit-based policies that are so common, and note once again a major finding from earlier Briefings – the difference between good strategy and poor can be huge! Here, we got our brand profitable in 12 months, whereas [ironically!] our profit-based rule in Briefing 69 took two years to generate profits, and the impact-based rule reached a higher monthly profit rate.
Flow-rate-based rules are especially important in the case of customer because these flow-rates are so hard to control, or even understand. As Briefing 16 pointed out, many flow-rates just do exactly what we want them to do [if you want more capacity, simply buy it] or nearly so [if you want more staff, hire them]. Other decisions, though, should still focus on direct impacts, rather than relying on ratio-guides or overall performance. How much training should we be doing? Spending X% of revenue on training cannot be a good rule, because it takes no account of what that training is doing – raising or maintaining skill levels.
Until next time…
If you would like to receive the series from the beginning in your email inbox, please register on the strategy Dynamics website and subscribe to Briefings in “My Account”
The perils of bench-marking
A whole industry has grown up for research organizations that survey firms in a sector and sell back to them anonymous performance rankings. Companies then compare their performance on certain issues with “bench-marks” — the best performance to be found among their competitors. Bench-marking may help you check that you are OK on key measures of operating performance, like quality levels or productivity, but it can be dangerous when used for making specific decisions.
One oil company found that its maintenance cost per unit of production was much greater than the “best in class”, and the gap represented huge potential savings. Sure enough, pursuing this benchmark reduced costs and boosted profits — for a while. Five years later, the company’s equipment was in such a poor state that breakdowns and emergency repairs soared. There were even worries about safety. Spending had to be raised far above the original rates just to stop things from getting still worse.
What’s wrong with benchmarking? Even if you are comparing yourself with similar competitors, the danger is that the allegedly best-in-class competitor may itself be making a mistake. And there may be subtle but key differences that make the benchmark incorrect – this oil company was operating an older and more complex range of equipment, so of course it needed to spend more on maintenance!
This briefing summarises material from chapter 8 of Strategic Management Dynamics, pages 541-545.
Read more about the book on our websiteShare