MAXIMIZING YOUR MARGIN - The Power of Accurate Cost and Willingness to Pay Predictions on Every Transaction

As a carrier, you know that it is complex to estimate the cost of your services and even more the maximum price a customer is willing to pay (WTP). Yet, as common sense tells us, these two factors play a vital role in setting your prices. Better cost prediction allows you to win price-sensitive deals at a lower but still profitable price; and better WTP prediction allows you to capture more value and improve your margin rate.

The objective of this article is to quantify the corresponding benefits.

But before that, let’s see how you can improve cost and WTP predictions when building an offer for a customer for a spot transaction or an agreement.

How to improve cost predictions

An effective cost model for pricing correctly predicts the incremental cost generated by a spot shipment or a shipping agreement. In practice, carriers use different types of models for this prediction:

• Simple models such as average cost per parcel;
• Activity-based models with multiple cost drivers (weight, volume, distance, pickup and drop-off densities, etc.) with different granularities (country, hub, route, zip code).

However, even sophisticated cost models can still be improved to be more effective for pricing by:

• Clearly differentiating between variable costs, capacity utilization costs and fixed costs;
• Calculating unit costs based on full capacity instead of used capacity;
• Being forward-looking (based on planned costs) as pricing is, rather than backward-looking (only based on past costs).

How to improve WTP predictions

Most carriers differentiate prices based on the size of the opportunity (in number of shipments or revenue) to take into account the negotiation power and the lifetime value of the customer. Some also differentiate by sector of activity. Additionally, all good salespeople instinctively assess WTP based on customer/ prospect characteristics, offer value, buying journey and competitive context. However, this assessment is most often based on gut-feel, is fuzzy, unquantified and subjective: ask two salespersons and you will get two different estimates.

A more robust approach is building an enterprise data-driven WTP prediction model relying on:

• Customer characteristics, offer value, buying journey and context;
• Price benchmark with similar deals/ customers;
• Win-loss analysis of similar offers;
• Market prices.

Such a model can then be updated on a periodic basis (monthly or quarterly) to complement salespeople's intuition and empower them in price negotiations.

Quantifying the impact on your margin

What would be the impact on your profit of improving the prediction of cost and willingness to pay when you are building a quote for a contract or a spot shipment?

To respond to that question we have defined a mathematical model of the pricing and customer decision processes and then developed the analytical formulas of win rate and margin for different scenarios of prediction errors.

We consider three market segments based on the size of the deal: Large (L), Medium (M) and Small (S) and assume the following value for wtp and cost per shipment.

It is assumed that wtp takes into account the market context, i.e. the value of competitive offers and their price. The objective is to measure the impact of the error of prediction of cost and wtp on the win rate and the margin.

The mathematical model and detailed results are described in the white paper (link at the end of the article).

We consider 4 different scenarios depending on the assumption of average error on cost and willingness to pay, as defined in table 2. The objective is to calculate for each segment L, M and S the average win rate and margin that will be obtained in each scenario and compare with the baseline.

The baseline (Scenario 1) represents the current average error performed by the carrier when predicting cost and willingness to pay. Note that these errors can actually vary depending on the type of quote (contract versus spot), by segment (L, M, S) and also depend on the maturity of the quotation process and the decision support data available. They can be estimated (more or less easily) based on historical data.

Overview of results

In global terms, after weighting the results of each segment L, M, S by the corresponding share of shipments, as shown in table 3:

• Reducing the cost prediction error from 20% to 10% (Scenario 2): increases the global win rate from 24% to 28% (+17%) with a slight increase of the margin. The captured margin increases by 20%.
• Reducing the WTP prediction error from 20% to 10% (Scenario 2): increases the global win rate from 24% to 28% (+17%) but also increases the avg margin by 22%. The captured margin increases by 42%.
• Combining the two effects will boost win rate to 33% (+37%) and captured margin by 64%.

Notes:

• The % Margin Captured is the most relevant pricing KPI of the quotation process.
• Improving win rate and average margin are not contradictory objectives because by better predicting cost and WTP, quote by quote, it is possible to reach both objectives at the same time.
• Reducing the error also reduces the probability of accepting transactions with negative margin.
• The results by segment (L, M, S) show some specific variations. For example, in the case of segment L-Large, reducing the error of prediction of WTP has in theory a very high impact on captured margin, but is extremely difficult to obtain in practice for these large deals.

As the impact on margin improvement of reducing the error is linear, table 4 presents the impact in case of reduction of error of 1%, 2%, 5% and 10%.

In conclusion: reducing cost and WTP prediction error by 1% increases captured margin by 6% (2% attributable to cost and 4% to WTP).

You can read the full study, including the mathematical model and the detailed results, in the white paper.