Advance pricing optimization

In retail, there is much more to pricing than just merely aligning all your prices to the lowest prices that can be found on the Internet. One of the most painful facts in global commerce is that there is always someone, somewhere, selling the same product as you at an unreasonable price. Naively aligning all your prices with your competitors is not a viable strategy. Lokad helps your commerce go beyond the “me too” pricing by implementing smart strategies where margins are thin when they foster growth, and where margins are not so thin when they do not hinder sales.

You can’t optimize your prices

In fact, only “pricing strategies” can be optimized. The distinction between “prices” and “pricing strategies” is subtle yet critical, and is too often overlooked. You prices depend on multiple factors: purchase price, return rate, competitor price … and all of these factors are changing all the time. As a result, price Y might be the good price point for January and a bad one for February. As underlying market conditions are constantly changing, it may often feel like it is not possible to perform rigorous and effective pricing experiments.

In reality, it is possible to perform rigorous pricing experiments, however, it is not the prices that are going to be benchmarked, but the underlying pricing strategies. A pricing strategy is the logic that helps crunch all the relevant input data for a given business in order to re-compute the latest applicable prices. Let’s illustrate how pricing strategies can be tested.

If you have, say, 200 products and 2 pricing strategies A and B, it is possible to use strategy A for half of the products, and to use strategy B for the other half of the products. Prices keep changing, possibly on a daily basis according to strategies A and B, however, after a few weeks, by comparing the respective results of A and B, it becomes possible to decide which pricing strategy is giving the best results.

Lokad is designed to give your business the possibility to craft your pricing strategies by starting with very simple ideas, and refining them to give very advanced suggestions as pricing practices mature within your company. With Lokad, your business can generate price lists, re-import those lists into the business systems you are using and visualize the results in order to assess which strategies are working best.

Theory vs Practice

In commerce, pricing tends to suffer from two opposing philosophies. On one hand, on the practitioner’s side, we have the “intuitive” pricing method, where choosing the “right” price is a manual process that relies on the manager’s intuition (and probably quite a few Excel sheets are well). As this type of pricing strategy only exists in the head of the manager, it is extremely difficult to scale this approach, and the results are highly dependent on the manager himself. Furthermore, when the manager leaves the company, a good chunk of the company’s valuable knowledge leaves with him.

One the other hand, on the theorist’s side, we have the “econometric” pricing, whereby choosing appropriate prices relies on advanced mathematics. Since Lokad is mostly composed of a (large) bunch of data scientists, statisticians and mathematicians, this approach was very tempting for us to adopt. However, years of experience have taught us the hard way that as far as pricing is concerned, “naive rationalism” generates endless problems. It is not because prices are optimized with a formula that this formula is “right”: pricing is a highly domain-specific issue, and detailed business contexts need to be factored into the pricing approach in order to obtain numbers that make sense. The naive use of econometric formulas just makes you accelerate in the wrong direction.

Building a pricing strategy

Lokad’s approach consists in letting your company build its own pricing strategy. This strategy deeply relies on the business insights that you have, and any software vendor that claims to have more insights than you do on your own business is probably severely delusional. Pricing strategies rely on experiments, typically A/B tests. However, since A/B tests are both time-consuming and short-sighted, the emphasis should not be on testing everything, but rather on the incremental discovery of a pricing knowledge base.

Lokad offers the possibility to gain understanding about how the market responds to prices with one hypothesis being tested at a time, and to systematically draw domain-specific conclusions from every experiment. By putting together all of these insights, it becomes possible to devise an overall pricing strategy that relies on deep expert understanding, and not merely on some black-box numerical optimization toolkit. In practice, a pricing strategy is a long list of simple insights such as:

• Don’t keep pressuring prices while you’re about to run out of stock
• Stop wasting AdWords on products that are not competitive for a first contact
• Prevent dead inventory from piling up by offering the right discount at the right time
• Stop doing promotions for products you would have sold full-price anyway
• Etc.
  However, unlike traditional pricing methods, Lokad places emphasis on a highly quantitative approach. Prices are not expected to be “good”: prices are expected to be “proven” to be “good”.

You can’t dodge “programmability”

As market conditions are changing all the time, your prices should change accordingly all the time too. Any approach that relies on manually editing individual prices not only suffers from the impossibility of being tested and hence measured, but it also suffers from dramatically low productivity. Therefore, the execution of pricing strategies should be vastly automated.

Any software that automates pricing strategies (shameless plug: we strongly recommend Lokad) needs to be able support just about any strategy you can think of. Otherwise, your business is stuck with a shortlist of pricing recipes that unfortunately are not necessarily tailored to your business. In order to check if a given software is “powerful” enough for your specific case, there is a very simple test - the “Excel” test, according to which a pricing solution should be capable of delivering any pricing strategy that could otherwise be implemented in Excel.

Indeed, if a software cannot do what can be achieved in Excel, is there really any hope to do anything smart as far as pricing is concerned? We believe not. At Lokad, we have developed a tiny programming language named Envision which is uniquely tailored for commerce. By leveraging this language together with the other resources offered by Lokad, we can support your company in building a great pricing strategy, driven by true expert insights and vast automation.