While many focus on the current economic uncertainties and ‘doom and gloom’ articles, I am witnessing a material increase in investments being made by corporations to move their business infrastructure forward. Simply put, there is never a bad time to maximize revenue – in a good market you want to capture as much as you can and in an uncertain market you want to squeeze as much as you can from the revenue you can get. Tariffs, and trade wars – these too shall come to pass and many companies are looking beyond that and trying to figure out the next evolutionary step for their business infrastructure.
Terms like Machine Learning (ML) and Artificial Intelligence (AI) feature prominently in almost every CIO’s project portfolio in 2019. We are finally moving beyond the over-hyped promises and disillusionment phase into practical implementations of these technologies. There are some very practical implementations of AI and ML when it comes to increasing and optimizing revenue driven through the channel.
It is always important to define the problem we are trying to solve before discussing how technology may be used to solve that problem.
The Problem: So Much Data, So Little Time. Solutions like Model N’s Channel Data Management have made Channel Sales and Channel Inventory data readily available on a weekly and daily basis with a very high degree of accuracy. This data comes with all the dashboards and reports to understand where the problem starts. Managers and executives who are pressed for time and need to make decisions are literally drowning in reports and dashboards. These typically fall into two categories – the high level directional information which tends to get used often. There are the specialty granular reports created when someone wants to drill down into a problem – those serve their purpose for a while but are rarely maintained and overtime lose relevance. It becomes an impractical effort to maintain and manage potentially dozens and hundreds of different drill downs. This is a huge missed opportunity.
High level trend reports keep track of KPI’s but they are not effective in helping managers uncover micro-trends and opportunities to improve. Here are a couple of examples:
- Sales Trends: Sales in Europe, Middle East and Africa (EMEA) are growing this quarter and forecast targets are met. This is a good report but it says nothing about what opportunities and threats exist in the region or the performance of specific product lines. What if the underlying data would show the reason sales is hitting the target is because specific channels are over-performing while others are under performing? What if the underlying data indicates a new product is really gaining traction while other products are trending down?
- Inventory Trends: Overall Inventory levels and rate of consumption of inventory are as planned. But what if three channels have slowed down their rate of shipping inventory?
The root causes for changing channel sales performance and/or inventory buildup is varied. Reasons include: Competitive pressure, overall economic conditions, change in channel personnel, competing suppliers buying market share with new incentive programs, training issues – the options are limitless and the corrective actions are varied. The problem is the information is buried under the high level dashboards. Nothing grabs the attention of management by saying “In this territory our channels are slowing down selling product family X.” Managers can always drill into the data and look for this information but when considering companies can have hundreds and sometimes thousands of products, dozens and sometimes hundreds of channels it is impractical and not sustainable to be in drill down mode to search for opportunities to adjust pricing, or introduce new incentives or invest in training in a specific channel, specific territory or specific product line. Most of the time, companies miss the opportunity to proactively engage in a timely manner and to change outcomes in real-time.
This is where Machine Learning and AI can really make a difference. Algorithms designed to investigate and find changes at granular levels like a specific channel location or specific product lines and extrapolate a predictive model of a business outcome, if emerging trends persist, allow machines to surface the opportunities to make changes (pricing, incentives and other) in a sustainable and repeatable way. Of course, these models can become more and more advanced when factoring in competitive information and seasonality. These models can also predict based on past performance opportunities to increase revenues by better aligning channels and end customers. Lastly, the real root cause analysis and recommended actions such as price change or the timing and model of an incentive program could be driven by ML and AI engines. While the latter part of this vision is probably still “bleeding edge” the basic idea that ML/AI can do a better job than humans to surface problem areas that require action is here and now. This use case alone can allow companies to drive 3% to 5% improvement in their revenue growth and revenue quality through their channels.
If you want to learn more about such opportunities and network with your peers in the industry, come to Rainmaker 2019 – see you there.