How to Measure Your Pricing Effectiveness with a Custom Pricing Index
[/color][/size][/b][/i][/u]
We were highly motivated to improve our pricing: we had key products with declining average sales prices, products with wildly varying discount rates, a competitor that was baiting us into a price war, a sales compensation plan that did not reward salesreps for good pricing behavior, no effective means of enforcing discount policies, and customers that were accustomed to paying the same price forever- it was not a good situation.
What did not work:Total New Revenue - this is ultimately what we were trying to drive (along with profits of course), so it seemed natural to use this as our key metric. But using a number like Total New Revenue or Total Revenue makes it impossible to tell if the results you are seeing are due to better pricing or higher sales volumes.
Average order size - this metric was better in that it controlled for overall sales volume, but it was heavily influenced by the product mix sold. So if we ran a marketing campaign for one of our higher-priced product lines we could misinterpret the higher average order size as being the result of our pricing work, rather than the natural effect of which products were sold.
What did work: a Custom Pricing IndexWe spent a couple of days researching the academic literature, online forums, and talking with peers at other companies, and we were not able to find a good solution. So we created our own Custom Pricing Index. Here is how it worked:
We identified a basket of products that represented the overall business - at least one for every important product line.
Every product in the basket needed to be sold in enough volume that we would have data for any time period.
The products in the basket needed to be picked with complete specificity - in our case we needed to specify both a product code and a quantity (# of users).
We created separate baskets for products sold in different countries. We could have commingled data across multiple countries, but found it easier to evaluate performance on a country-by-country basis.
We then pulled sales data for each of the products in the basket over the last few years, bucketed by the time periods of the sale (e.g., month/yr or quarter/yr). This data extract was harder to execute than it sounds, as we had a bunch of duplicate contracts data to wade through, we had a bunch of orders where one product was "free" (we spread the discount across all products on the order), and we had a bunch of products that had changed names over time (we created a synonyms table) - hopefully, your data will be in better shape.
We then took the average sale price for each product in the basket and added them together to create a price-weighted index in much the same way that the Dow Jones Industrial Average is calculated. We considered weighting each product based on the number of units sold (more like the S&P 500 index), but we ultimately felt like the understand-ability of the simple-addition calculation was worth more than the added precision of varying weights would provide. If you do choose to implement a volume-weighted calculation, it is critical that you keep the weightings consistent over time - if you recalculate the weighting in every time period, you will not be able to separate the pricing behavior from the changes in product mix sold.
To visualize the results, we created a simple graph and table showing the performance of the index over time. The following is a sanitized example.
We were then able to use the trend data to set targets for improvement and added this to the agenda for regular marketing performance review meetings.
There are a couple of valuable features of this Custom Pricing Index:
It is easy to understand. I refer to it as "like the Dow Jones" and people instantly get the concept.
It can be calculated using historical data that you likely already have - this is not the kind of metric that only works going forward from the time you decide to start measuring it.
You can easily change the components of the index over time by adding a divisor to the calculation so that the index remains stable before and after the product basket changes (in the same way that the Dow Jones divisor gets updated when a new company is added).
It provides a quantifiable way to measure the results of pricing work, independent of the volume or mix of sales.
Reference:
https://www.linkedin.com/pulse/working-how-measure-your-pricing-effectiveness-custom-jim-o-leary
