7.9. Policy Design, Growth and Dynamic Complexity

The dynamical issue at the heart of the market growth model is the coordination of production capacity with sales force and customer orders. The model shows that it is all too easy for companies to under-invest without realising it. By assumption, the market opportunity is large and, at first glance, it would seem impossible to fail. Hire a sales force to tap demand and build a factory to supply the customers. In reality these actions are interdependent. A management team cannot build a factory and invest the huge sums of money required without solid evidence of demand. Customers will not buy a product unless it is available, no matter how attractive it is in other respects. The marketing department cannot hire sales people without revenue, at least not in the medium to long term. Because of these interdependencies, many factors influence business growth and successful market development. Success depends on the importance customers attach to quick delivery, their patience and perceptions, on the size of the sales budget, how much sales people are paid, how quickly they can be hired, on the operating goals of the factory, the business pressures that drive capital investment, top management attitude to risk, the speed of adjusting production capacity, the flexibility in utilisation, and so on. In theory it is possible to find a combination of all these factors that ensures success – optimal growth if you like. But the practical challenge is more pragmatic – to discover those few factors that avoid stagnation and decline.

The secret of growth in this case is coordination between sales force, capacity and customer orders. The previous simulations show that sales tend to grow quickly when delivery delay is low. What does it take for the firm to achieve this condition?

We can use our knowledge of the feedback structure to make informed guesses and then test them with simulation. Delivery delay arises in the interaction between the sales growth loop and the capacity expansion loop. If capacity expansion is well matched to the expansion of the sales force then delivery delay should remain reasonably stable. Capacity expansion depends heavily on the goal formation process for delivery delay operating goal and on the investment bias applied by the top management team. The more pressure these processes bring to bear on investment the more effective is the balancing loop for capacity expansion and the more likely it is that capacity will keep pace with growth in sales. For these reasons, both delivery delay bias and delivery delay weight are high leverage parameters for sustainable growth, as earlier simulations have already shown. One might suspect therefore that combining optimistic investment (delivery delay bias = −0.3) with a well-chosen fixed operating goal (delivery delay weight = 0, delivery delay management goal = 2) will unleash even more market potential. The way to find out is to simulate, which the reader is invited to do.

The policy changes proposed so far fix the coordination problem from the supply side, by boosting investment. What can be achieved on the demand side? Our analysis suggests, paradoxically, that the firm fails to grow because its sales force is too good at winning orders – too good relative to the firm's willingness to invest. So an alternative policy might be to reduce the strength of the sales growth loop, say by reducing the fraction of revenue allocated to the sales force. In the base case this fraction is set at 0.1 (10 per cent). Again the way to test the policy change is to simulate. Other policy changes are possible too. For example, one might test the effect of reducing the time to adjust the sales force or increasing normal sales force productivity. Combination supply side and demand side changes can also be investigated, though complex parameter juggling is best avoided because the corresponding real world policy change may be difficult to implement and the marginal performance gains small. An interesting structural change is to make product price a function of the order backlog. These and other experiments are left as exercises for the reader. Remember the objective in policy design is to find high leverage parameters that result in significant and robust performance improvements.