In today’s volatile business environment, many firms look to optimizing supply chain and seeking to secure additional sources of supply of scarce materials. However, one extremely influential group is sometimes overlooked: the role of the salesperson can be huge in determining the “right” projects to serve in times of uncertainty.
With much focus on minimizing cost, optimizing processes and streamlining channels to reduce the cost to serve, sales teams can also play a major role. Their technical-commercial skills mix has long been a scarce resource to be treasured, nowadays a data-informed view that helps them spend their time wisely has never been more relevant. The combination of the sales professional’s market and client knowledge, combined with data-driven insight into deal and project profiles, helps organizations pick the right Deals to serve, especially when supplies are limited. The right deals could be those with the highest margin, or with the most future potential, strategically and financially. An experienced human sales brain can compute many ‘deal’ parameters to make the right choices; supported by data science it can make even more.
So, in times of supply chain shortages and rising costs, how do companies know which are the best deals to focus on?
Why Opportunity Scoring?
Opportunity Scoring helps business prioritize the right deals, depending on the company’s priority, often it is those with the highest margins and the highest win chance. Especially, when it comes to high value project-based deals, where you incur a high cost of sales to compete, where you need to invest scarce resources upfront to demonstrate your capabilities, you want to be sure you are investing those resources in the right place at the right time.
With so many moving parts in global operations, and many deals to lead at once, salespeople often don’t have enough time or information to properly evaluate every sales opportunity according to a standard process. Most of the time, they rely on their business judgement and previous experience to guide their decision-making. But, in such pressured circumstances, there is the risk they could lose deals and projects they were hoping to win, resulting in lower revenues and win rates.
This problem is compounded in manufacturing, where many very experienced sales professionals have been in the industry a long time, often building up their deep and intricate knowledge of the market, customer, “perfect project” during a whole career in a certain company or sector. Their knowledge is akin to gold dust! Many of these salespeople are likely to retire in the next 5-10 years, together with their knowledge, unless we codify their insights, helping newer or less experienced colleagues along the way.
Opportunity Scoring, using machine-learning modelling, gives the sales team a second, data-driven opinion on which deals are most likely to be won and helps them to organize their time and resources accordingly. Rather than relying solely on their business judgement and experience to prioritize a deal, they reinforce their decision making, and challenge their hypotheses to improve the decision-making process throughout every part of the sales engagement. They can focus on the deals they are most likely to close, or they can develop a new sales strategy for those deals that have a lower score than they would have predicted. Over time, “battlecards” are created that help teams break into traditionally difficult deals, such as competitor stronghold accounts.
Here, new, or less experienced colleagues benefit from this collective knowledge as the opportunity scoring model essentially ‘codifies’ the knowledge gained over the years and puts it to work for the whole organization.
Opportunity Scoring provides you the benefit of “codifying” the collective brainpower and decision-making minds of the company’s top salespeople. Using machine-learning modelling, the model learns what success looks like, transforming the deal characteristics and information that affects decision-making into a score and insights that teams can use to inform their sales strategy, which becomes available to everyone in the firm. As a result, the company’s “sales expertise” is preserved and enhanced, and continues to improve as sales teams see the benefit of the second opinion, encouraging them to input more deal data more frequently, continuously improving the model’s accuracy in doing so (as well as keeping the master data stewards happy!).
Further, sales directors are able to use opportunity scoring at an aggregated level to help forecast their likely sales performance against target. They can use the scoring to identify and inspect must-win deals at risk, secure resources and coach their sales teams to focus on the right deals they want and are most likely to win.
What is our approach?
To help sales teams qualify their deals more easily, we first try to understand their existing qualification process and specific business goals. Using our library of machine learning (ML) assets, we configure and train our ML model to learn what successful deals look like, based on thousands of historical deal profiles for which the outcome is already known. The model is used to score open opportunities based on their likelihood of being won. This enhancement is fully embedded into the sales team’s existing process, such as a dashboard or CRM system, depending on how they work. It is intended as evolution not revolution, ‘simply’ offering a small improvement in the form of a decision support tool, to help them make data-informed decisions.
Once we are all happy with the model’s performance, and sales teams feel that they can “trust” the model, it is integrated into the sales team’s systems so that it can score opportunities live and publish the results in the salesforce’s CRM. Through a dashboard embedded directly in the CRM tool, sales teams can attain a ‘second opinion’ on their deals’ likely success. Additionally, the sales leader can look at all sales teams’ deals together, and get a ‘second opinion’ on their overall forecast.
What are the benefits?
One of the biggest benefits is the ability to prioritize the “right”, usually the highest margin deals. In current times, prioritizing which deal would be most advantageous for the company is often more important than overall deal revenue. We generally see win rates increase by 1-4% in mature markets over a year. Most sales teams report a sense of order and relief to be the biggest benefit, as they were able to concentrate resources – and crucially get backing from other teams – on the deals most likely to be won, rather than spreading themselves and their teams too thinly across many deals that they felt they should be pursuing.
Finally, data-driven predictions assist teams to improve their Funnel Accuracy. Not only do teams see the benefit of entering more data more often, but they also question and correct data for deals with lower scores than they expected, where they may find data inaccuracies. Once fixed, overall data quality improves, as do the scores the model is able to provide.
Our experience
We have already teamed up with several global manufacturers to optimize their data and help them to prioritize the right deals.
In one such case, the sales teams needed help in identifying the right priorities and chasing the deals with the highest potential. A new CRM system additionally increased the workload of the sales organization by raising the reporting requirements. Starting to implement data-driven solutions, leveraging the reported data, and demonstrating the value in capturing and analyzing this information for the sales organization was a logical next step.
Amplify Analytix created and deployed а predictive machine learning model to calculate opportunity scores, which indicate from 1- Very Low to 5-Very High the potential chance of winning an opportunity. The score is (re-)calculated at two moments of the lifecycle of an opportunity – at its creation in the CRM and when the first offer to the customer is submitted. The calculated scores are visible in the CRM and allow the Sales Reps to evaluate their winning potential and estimate the effort necessary to close the deal.
By bringing the benefits of machine learning in an easy way to the team, we helped our client achieve greater success in a variety of sales areas. The solution led to an increase in efficiency for the salesforce by identifying the high potential deals, enabling them to prioritize resources better.
Machine learning models were used to achieve the right opportunity scoring. During the “Modelling“ stage, two gradient boosting algorithms were developed to predict the probability of an Opportunity being won in the future. The first model evaluated the opportunity’s potential at the time when it was just identified and created in the CRM. The second model evaluated the chance to win the opportunity after the first offer has been submitted to the customer. At this moment, a lot more details are available about the specific deal which are used as predictors for the model.
The model was created and deployed in the Azure cloud environment using the following infrastructure: the creation of a new Opportunity entry in Microsoft Dynamics (MSD), the API sends a message to the Azure Service Bus (ASB).
The message waited in the ASB queue to be read by a Virtual Machine (VM) set up at Amplify’s Azure Active Directory. When the message was received the VM called the Dynamics API got all relevant data for the newly created record. The data then was cleaned and transformed into a scoring-ready format. The VM calls upon the algorithm to score the ready data. After the data had been scored, a call was made back to the MSD API to populate the “Opportunity Score” field for the given record with the newly calculated score value.
The model was trained on CRM information on the opportunity and its customer such as the equipment type to be sold, the value of the deal, the past relationship with the client, etc. The quality of the data in the client’s sales organization has improved because of the implementation of the CRM.
In the end, our client increased conversion rates of sales opportunities by 3%, reduced the inaccurate data in the CRM system by 50%, boosted salesforce efficiency with 1000+ better-utilized working hours, and gained more than €1,5M additional revenue in 3 months.
The custom Opportunity Scoring models we create are much more than a built-in software solution. They are suitable for all interested parties, especially for the companies which are working on improving their data quality.
We carefully take our customers on a joint step-by-step discovery data journey facilitating their full involvement at every project stage, collaborating and enabling us to be on the same page throughout the entire process. We help break down the numbers, trying to explain any context and actionable insights that come out of the black box data model. We work with our clients, and they learn with us. As a result of becoming familiar with the behind-the-scenes process, they are more likely to believe the results and build a sense of trust and reliability. Finally, they gather new knowledge – they understand how the discovery was achieved, and how we arrived at the insights. In other words, we don’t just give them the fish, we teach them how to fish!
In the end, the client will not only receive a model that helps prioritize the right deals, but they will also have the opportunity to build a data skills foundation across traditionally less data-literate teams, which will bring huge dividends when it comes to data-driven decision-making.
At Amplify Analytix we have more than 15 years experience at big corporations, where we had the chance to see the real challenges of sales teams from the inside. This taught us to use data insights to get a comprehensive business perspective that is otherwise missing. We are experts in the implementation of advanced data science solutions in the Sales domain which don’t require any client’s internal knowledge or people resources’ investments – they only harvest the benefits of their business!
In these times of volatility, Opportunity Scoring can help B2B Manufacturing companies get back on track and focus on those deals that will provide more benefits. By bringing the advantages of machine learning in an easy way to our clients, we help them achieve greater win success and increase their sales productivity.
