It’s been nearly 15 years since “big data” broke into the business lexicon, often upending long-held assumptions and elevating the critical role played by data scientists and suites of analytic tools. In that time, most of the attention has been on the B2C world. Only recently has the B2B sales world started to appreciate the scale of the benefits data-driven sales can deliver, from prioritizing leads to enhancing existing relationships with customers, by offering the right product at the right time for the optimal price.
The limited experience B2B businesses have with analytics puts them at risk of making some of the same mistakes their B2C colleagues suffered through. Going into an analytics program with only a vague sense of objectives, becoming enamored with specific technology, or building with scant concern for how tools will be used in the field are all likely to lead to wasted resources and poor results.
Analytics isn’t likely to identify a whole new way to do business; rather, it will provide insights and opportunities to sell more and do so more efficiently. For more than two years, we have helped a number of B2B companies harness analytics to create significant value. We have seen the speed of the initiation of first sales increase by 50 percent, churn reduced by 25 percent, sales from new accounts rise 10 percent, and a 2 to 5 percent return on sales through pricing. Along the way, we’ve identified five critical lessons to ensure big data serves your business’s needs:
1. Focus on clear business objectives—and ignore shiny objects
We’ve seen it time and again: too many sales organizations start their analytics implementation by asking what tools they’re going to use. It’s easy to fall into the trap of bringing in an off-the-shelf solution favored by an operating unit or key stakeholder. And sometimes it does pay off, but only in the short term.
Instead, companies need to keep their eye on the fundamentals of the business and what problems stand in the way. Start analytics implementation by defining those challenges, and then build internal knowledge through testing and iterate with tools on hand. These trial runs will build a case for support from top management, and that’s essential not only for budget allocation but also for gaining an organization-wide mandate to go further.
Then, and only then, evaluate analytics tools.
2. Help your sales team “trust the data”
An international chemicals manufacturer plunged headfirst into an analytics program, convinced that the answers to how to sell better were in the data. Unfortunately, the sales team wasn’t involved from the outset. Not knowing the inputs into the program, they didn’t trust the recommendations, and they gave the program’s findings the cold shoulder. The sales reps were certain they knew their customers well, and they trusted their instincts over a black-box analytics engine.
Overcoming sales-team skepticism requires a dedicated approach to building trust. We have found four elements to be important to success:
Create transparency. By providing transparency into how the algorithm is built and how insights are derived, companies are much more likely to persuade salespeople to trust the analytics. Take time not only to provide the results of the data analysis to the end user but also to offer final recommendations to the sales force and confirm that they have confidence in them.
Involve your salespeople. It’s essential to work with the sales team to determine what they really need. This goes beyond standard adoption protocols for releasing software or tools. The best analytics teams work with sales reps as partners and evaluate solutions, such as ways to improve their relationships with customers, from their point of view. Look for pain points that analytics can help solve—for example, better pricing options or a simpler reordering process to improve customer service. Learning from its earlier mistake, the chemicals manufacturer held workshops with sales reps to map their typical workflows and “user journeys,” including common motivations, goals, and pain points. The findings made clear which features would be of greatest value to the sales force and how current systems were failing to meet their needs. Working with the designer, the company went back to the drawing board and built a completely new minimum viable product (MVP) prototype customized to the specific needs of the sales force in various markets.
Start simple. Even the simplest analytics programs can uncover insights, such as underlying inefficiencies in market structures across suppliers, distributors, and customers. These insights can then drive significant changes in how the organization engages with clients. Start simply, and ramp up on the back of small wins (improved funnel conversion, for example). This helps sales reps become more comfortable with the insights and understand how data can direct their decisions.
Show the value. Sales reps ultimately want to sell more, so be clear about how analytics can help them do that. The best tools can give salespeople and their managers a window, for example, into individual performance against targets, show where there are performance gaps, and then identify specific opportunities to close those gaps while providing concrete recommendations on pricing.
After the chemicals manufacturer reworked its analytics program for greater transparency, the program identified cross-sell opportunities based on what clients who were similar in key characteristics such as size, sector, and location had purchased, and the sales reps could see the logic behind the recommendations. When they acted on those findings, not only did sales grow, but customer satisfaction increased around 25 percent for those accounts affected. The company realized 4 percent above-market sales volume growth.
3. Make it easy to use
Even the strongest insights can only translate into measurable impact if sales teams are able to act on them. The best teams use design thinking to develop tools that put sales-rep experience at the center of the process. That means developing tools that are simple to use, delivering information that’s easy to understand, and providing insights or recommendations that are easy to act on. Make sure that the insights you want to make available are convenient to access and easy to understand for both reps and managers. The insights generated by analytics, for example, should be fed into the tools the sales force is currently using—no one wants to learn how to use another tool. Dashboards must be simple to understand and clear on recommended actions.
4. Start with the data that are easy to get
Combining data to create a perfect data set can be frustrating. That fact becomes painfully clear when an analytics program is trying to run across multiple systems that typically don’t communicate with one another. This is where too many analytics efforts fail, incurring delay after delay as they try to make the combined data perfect.
Call it ruthless pragmatism, the 80 percent solution, or common sense, but experience has shown us that successful programs start with the data that are easily accessible in one system or in systems that are already communicating well with each other. Now is not the time to seek out third-party information and invest the time necessary to negotiate access and merge the feeds.
Only the data most relevant to the business challenge at hand should be cleaned and linked in a single data mart. These steps can be manual, as the only goal should be to prepare a first version of a data mart that can feed into the algorithm. This might include, for example, collecting the data via a simple CSV data file export or a quickly set up open database connectivity (ODBC) protocol. By focusing on speed and pragmatism, the first version of a data mart can usually be built in a few weeks, sometimes even just days. In fact, if this phase takes more than four to six weeks, you’re doing something wrong.
Go through the process again and again, build up the data sets, and refine the algorithms being developed until the program runs successfully every time. Use test-and-learn pilots to identify and target new customers, accelerate pipeline growth, and improve salesperson and channel conversion rates for a specific product group. Then, and only then, automate the process.
5. Build a team mind-set
Building a successful analytics program often means removing long-standing silos of data and analysis. For example, we typically see that companies can overhaul their sales-pipeline approach only after breaking down cross-functional barriers between sales, marketing, and their product teams. Typically, each team has its own tools and sources of data. This creates multiple blind spots. To overcome this, functions need to form an integrated team to share data and improve sales analytics.
To get there, a particularly effective technique is to launch a series of test-and-learn pilots to identify and target new customers, accelerate pipeline growth, and improve salesperson and channel conversion rates for a specific product group. With each “win” (goal achieved), the cross-functional team integrates additional existing internal data sets across customer relationship management (CRM), enterprise resource planning (ERP), and relevant product streams and can then enrich that with external data on competitors, intent, and other signals. This provides a richer capability, which makes adoption easier.
Such joint analyses can lead to the development of predictive analytics programs that, for example, identify the highest-propensity microsegments and inform the messaging to those microsegments. Leads are then prioritized and allocated to sales channels based on both value potential and customers’ buying preferences.
Throughout the process, companies usually find that addressing the general mind-set and working style of the people involved is critical to success. That includes providing immediate feedback and developing incentives that reward use of data. Measuring success by the joint team’s progress rather than by each individual function helps bond the team. The test-and-learn approach allows more agility, which is a great chance to spur a greater willingness within the organization to innovate.
B2B companies have been slow to embrace all that big data has to offer, often because they’re unclear about what’s possible or are intimidated by the complexity. But those competitors that are ready to move are empowering their sales team with insights that will translate to the bottom line.