BI, BA and BO but in a good way…
Business Intelligence, Business Analytics and Business Optimization
If you examine the pricing associated with product life cycle pricing, you are immediately overloaded with all the different aspects of a the life cycle to manage. However, if we break down the complexity, we find that the techniques draw on Business Intelligence, Business Analytics, and Business Optimization. There is a combination of all three tiers over time. While you may model a new product introduction, but then to market. At this point, it is about evaluating the pricing strategy. This may incorporate a price testing framework, demand forecast models and performance metrics. This in turn would require different resources to evaluate the outcomes, perhaps as part of a new product team. This is Price Execution.
Price execution incorporates people, process and necessary technology to unify different groups to facilitate the sales process, pricing process and drive supply chain processes. After all the price modeling, sales process modeling, and technology investments, it becomes time to go to market. As in the case of a product’s complete life cycle, the resources begin with product development, costing experts, marketing and sales. This then turns into sales, pricing analysts, contract management and financial support. When it is time to remove or sunset products, a different team will perform the duties.
Very few firms have a complete plan from sunrise to sunset. The concepts of price execution include,
- Business Process Modeling
- Contract Management – contract negotiation
- Contract Execution
- Identify & save changes to pricing
- Monitoring contract performance and adherence – BI reports
- Drive follow-up actions – reduce/remove profit leaks
- Demand forecasts, Pricing, supply chain
- Approval Workflow – contract changes, pricing changes
While processes can remain stable for many years, the underlying analytics can become stale due to changing market conditions, new competitors, catastrophic events, black swan events, etc. There is a growth curve to incorporating analytics. For some customers, they do not need the latest and greatest analytical options. On the contrary, it may prove detrimental because they do not have the resources to support the path to analytics excellence and then they are left with a system that they cannot maintain nor use.
Codarex helps our customers get ahead of the curve will minimal cost expenditure. We can provide the resources to evaluate your sales contract quoting tracks, the scientists to evaluate and redesign your pricing models, forecasting and decision optimization models. Understanding where as a company your lie on the analytics path is helpful to understand the next steps to take. In some cases, having an advocate to present your analytics case to software vendors can prove helpful in two ways: reduce your cost by ensuring the software meets your user needs and you get what you need without know what you need! Your advocate will ensure that your future needs are met.
The Growth of Analytics
Codarex has participated in the analytics evolution upheaval. provides analytical support services for sales, marketing and pricing applications. This includes contract negotiation and contract management. Furthermore, for many companies Marketing is the owner of pricing functions. Hence, we have provided support of segmentation, consumer choice models and price testing.
Descriptive Statistics
A business intelligence and reporting installation will publish descriptive analytics about your data. You can slice and dice the data across many dimensions. In essence the system describes the data by presenting, organizing and summarizing data in reports and dashboard widgets. Mean, standard deviation, rank, percentiles, frequency plots, Pareto plots, box plots, etc. are examples of descriptive statistics used in reports. Lastly, this is usually a “look back”, as history is summarized.
Inferential Modeling
Inferential modeling is the statistics of studying a random sample of data to understand the population. Typically a sample is used to create hypothesis tests, perform estimation, determine relationship among variables, make predictions, as well other modes. From the sample, we draw conclusions to drive decisions or make predictions about the entire population with a certain confidence. Price testing, product offer conversion, maximum likelihood estimation, ANOVA, promotion campaign analysis are but the surface of the many avenues of testing.
Predictive Modeling
Predictive analytics aim to predict what is going to happen and aren’t valuable unless they are actionable. The tools employed are forecast modeling, simulation studies, survey analyses, data mining and machine learning techniques. Examples of popular predictive analytics use cases include churn prevention, demand forecasting, fraud detection, and predictive maintenance.With the example of churn prevention, the goal would be to figure out what the customer is ultimately going to do and when so that the organization can intervene and hopefully avoid the churn (or at least mitigate the risks associated with it).
Predictive analytics has its roots in the ability to “predict” what might happen. These analytics are about understanding the future. Predictive analytics provides companies with actionable insights based on data. Predictive analytics provides estimates about the likelihood of a future outcome.
Optimization Modeling
As the name implies, optimization modeling pertains to solving large problems for an optimal solution given a bucket of constraints and rules. Optimizing price for a new product, optimizing list prices, optimizing the number of aircraft seats to save at a certain price, are all examples of optimization models employed in pricing. The need for a optimization is usually driven by the size of the problem in terms of actionable outputs, or managing many complex business constraints, and/or both. Linear optimization, quadratic optimization, mixed integer optimization, dynamic optimization and stochastic optimization are common models employed in control, pricing and sales.
Prescriptive Modeling
Prescriptive analytics is a new buzzword to address what firms have been already doing for years, namely, the combination of inferential, predictive analytic outcomes with optimization models, and other tools. The fundamental difference is that a predictive model may not address a business problem exactly. A prescriptive model is created to leverage the outputs of predictive and inferential models, to produce pathways to optimally attain some business metric. For example, a predictive model may identify a stock for a reasonable gain in the next 3 months. However, the model does not tell you the strategy to employ to maximize your return and minimize risk: should you buy and hold the stock; should you buy the stock and near term in-the-money options; or should you buy near term in-the-money options and sell near term out-of-money options. Prescriptive analytics quantifies the different paths to the predicted goal. Prescriptive analytics not only predict the future but suggest the actions to perform today to maximize some future measure, like consumer retention, market penetration, promotional pricing offer frequency, etc.. Prescriptive analytics attempts to quantify the effect of future decisions in order to advise on possible outcomes before the decisions are actually made. At their best, prescriptive analytics predicts not only what will happen, but also why it will happen, providing recommendations regarding actions that will take advantage of the predictions.
Prescriptive analytics take predictive analytics one step further — not only do they provide new information to make the aforementioned forecasts and predictions a reality, but they represent a paradigm shift and further model development. If predictive analytics cover what is bound to happen, prescriptive analytics aim to deduce the steps that should be taken to achieve a certain outcome — they’re much more actionable than their predictive counterpart. Staying with the churn reduction example, prescriptive analytics involve figuring out how to make the customers stay (such as by building targeted marketing campaigns for those customers like in this uplift modeling example).
It is no surprise that prescriptive analytics has had an impact in sales and pricing. Whether you build models utilizing machine learning to understand the complex relationship among attributes during a sales contract negotiation, or you lever AI to suggest the optimal number of contact points by contract type, these are examples of how prescriptive analytics can impact every step of a sales process. It is a new paradigm whereby only the best pathways are identified and suggested to internal resources to achieve corporate sales targets and goals. The holy grail is to incorporate optimal pricing to achieve conversion rates and profitability. Prescriptive analysis should be a goal of every major sales department going forward. The level of insights that can be gained into customer and sales rep behavior can literally be a game changer.
Price Execution: Rationalization
In order to keep you price execution on target, and keep ROI high on your investments in pricing software and quoting software, you must “oil the gears”. This is general maintenance to important master data, than rarely gets done. Codarex calls these rationalization projects. These projects are typically associated with dimension data such as product, customer, geography and other dimensions. This data can become stale for several reasons such as, market change, company mergers, slowly-changing dimension data, and fundamental business changes. In a recent engagement, we proposed an adjustment to the company’s territories because there was considerable connections between two adjacent territories. This was something they had not considered.
Rationalizing a customer’s out-of-control product hierarchy involves understanding the performance and associated network of SKUs. This would be including analyzing a products performance in terms of revenue and profit contribution, product complementarity effects, product cannibalization effects, bundle programs, and profit leakage. Companies rarely cleanse their product hierarchies, let alone, establish related product links. Hence there are hidden interactions with real revenue and profit implications. Without these linkage graphs, a price change in a single product could affect revenues from other sources.
In addition to product, we provide customer hierarchy rationalization. This can be in conjunction with a loyalty program review process. The customer – company relationship changes over time for various reasons such as competitive pressures, market pressures and behavioral pressures. To complicate matters, a customer’s interaction with your entire product line will change over time. This depends on the complicated relationship that your customers engage. Your customers may wish to reduce supply chain risk and use you company as primary supplier on some products, secondary on others.
Lastly, we have helped companies determine the optimal geographical segmentation. This involves understanding the impact of revenues across borders both locally and global. Geographical associations are typically aligned with revenue, currency, and tax implications, however, with advancement of enterprise software capabilities many business reporting functions are available natively in the ERP. Hence a company can now expand (or conversely contract) how they view North America, South America, EMEA, and Asia.
These exercises will determine how you handle sales and pricing which in turn determines how you will collect underlying data for contracts, and order fulfillment to create views on revenue and profitability.
Figure 1.2: Jackson Chou, Thesis, “Efficient Product Rationalization within a Single Product Portfolio”. 2013
Contact us to review your current pricing processes and assess what needs to be reset, what needs improving, and what needs replacing.