Case Study

Shopify Plus

Shopify generates 30% more sales with predictive lead scoring‍

Shopify Plus is the leading cloud-based, multi-channel commerce platform designed for large to enterprise businesses. Merchants can use the software to design, set up, and manage their stores across multiple sales channels, including web, mobile, social media, marketplaces, brick-and-mortar locations, and pop-up shops. The platform also provides merchants with a powerful back-office and a single view of their business.

Born out of Shopify, Shopify Plus launched in 2014 to provide high-growth, high-volume merchants a customizable solution without the need for large investments of time, money, or resources. Shopify Plus not only eliminates costly builds, crashed sites, and infrastructure that burns through development resources, it enables companies to scale through ecommerce’s largest network of support and award-winning Partners.

The business challenge

Shopify’s core platform powers over 500,000 merchants and attracts a high volume of leads. However, Shopify offers a wide variety of plans starting at $9/month. This results in a range of inbound leads ranging from very small businesses to multi-million dollars enterprises such as Nestle. The sales team at Shopify Plus offers enterprise-grade solutions for high-volume merchants, and large businesses. Identifying qualified leads for the enterprise solution from the large pool of leads received, proved to be a challenge.

Before implementing MadKudu, Adam worked with his data science team at Shopify to build a predictive lead scoring. The project was meant to help the Sales team focus on leads that had the highest potential for the plans they offered. Using firmographic data along with marketing data extracted from Hubspot, they built a model that accurately identified the top 10% of leads that would generate a third of the revenue.

However, Adam had heard about other companies implementing MadKudu and their success prioritizing leads based on the MadKudu scores. He prepared to run a blind A/B test where 50% of the leads would be scored and prioritized based on his team's score and 50% would be scored by MadKudu. The reps wouldn't know where the scores originated from.

How MadKudu helped

It only took Adam a few clicks to connect MadKudu to Hubspot to start the process.

The MadKudu platform started pulling conversion information about Shopify's historical leads. MadKudu automatically started enriching leads, deriving features (aka potential predictors) to predict conversion. Using advanced machine learning, MadKudu identified which combinations of features were highly predictive of the potential of a lead and an account. The results were staggering and showed that MadKudu could identify the best 4% of leads that accounted for 34% of the conversions.

When looking into the scores of leads that had been worked by the sales team historically, MadKudu found there was a high potential to improve the team efficiency by aligning their efforts against the predicted potential of leads.

MadKudu Leadscoring in action

Leveraging MadKudu's advanced machine learning, Shopify was now able to identify "A" leads which converted 42 times better than the "D" leads!

The results

Adam and the Shopify Plus team set up and ran a 2 month A/B test. Leads would randomly be scored either by MadKudu or by Shopify's internal model. After the test was finished the results were unequivocal,MadKudu outperformed the Shopify model by 30%

The benefits of using MadKudu were two-fold. It enabled the sales team to perform even better while working fewer leads. More importantly it allowed the data science team to increase their leverage by focusing on building intelligence for the Shopify app, improving the commerce experience for merchants across the platform.

Better alignment

MadKudu enabled the sales team at Shopify to spend their time and efforts on leads that carried the highest potential