Skip to Main Content
Technology

RevOps: The Missing Link Between AI Models & Business Value

Author
Sean Backe Director
Author
Sydney LaRose-Alexander Group, Inc.
Sydney LaRose Consultant

Companies are making substantial investments in AI models, data science teams and cutting-edge AI platforms with the hopes of gaining a competitive advantage. However, many CEOs and other executives across the C-suite find themselves questioning the return on investment (ROI) of these AI initiatives. The common issue is that these models are often never deployed, leading to unrealized value from AI investments. 

The fate of AI models, particularly in the go-to-market (GTM) context, is that they are often underutilized. Recent Alexander Group research showed that the top reason companies aren’t investing in AI is due to a lack of relevant use cases (reported by 83% of companies).

While AI tools can offer powerful insights, they still need a team behind the scenes to add business value, particularly when it comes to:  

  • Identifying relevant commercial use cases
  • Choosing the right training dataset
  • Validating model outputs pre-deployment 
  • Ensuring clear tactics for application

Revenue operations (RevOps) teams play a critical role in transforming AI algorithms from expensive shelfware to key drivers of business value creation. By closing the loop between AI models and GTM priorities, RevOps creates value by increasing the likelihood of successful AI deployments.  

The Role of RevOps with AI 

RevOps is essential for realizing the full value of data science by translating business goals into actionable steps to deploy AI models. Data models are often deployed using a standard process (known as “CRISP-DM”), which breaks down into six stages: 

Figure 1: Data Science & RevOps Responsibilities Across the CRISP-DM Process  

As shown in figure 1, RevOps is actually responsible for more activities in deploying AI models than data science teams. This is due to the “translator” role RevOps plays in tying business objectives to KPIs and then making sure the correct data is handed over to data science teams. After the data science team executes the modelling, RevOps jumps back in to ensure the model outputs make sense in the context of the business. Additionally, RevOps is also responsible for integrating outputs into other tools, creating automation around sales and marketing plays and training the sales organization on how to use the model outputs.  

Without RevOps, these AI algorithms are just powerful shelfware. In order to make this partnership work, data science & RevOps teams need to align on use cases, data and roles.  

Clear Use Cases

The key to unlocking this collaboration between RevOps and data science teams is alignment around specific go-to-market use cases. Alexander Group has identified 20 different proven use cases for AI/ML that drive value in a go-to-market organization (see figure 2). Notably, these use cases not only touch every stage of the customer lifecycle, but their requirements also span from simple machine learning to complex generative AI models. Whether your primary pain point is demand generation, customer churn or anywhere in between, there is a relevant AI/ML use case that can drive value. Likewise, whether your firm has a mature and tenured data science team, or your company’s AI journey is just getting started, there are steps that can be taken to create value with the resources and technology already available.  

Key Topics
Close
Subscribe to our latest insights

Connected Data 

RevOps and data science teams share common interests when it comes to data management. The mandate of RevOps is to create coordinated growth plays across marketing, sales and customer service. This inevitably requires bringing together data from diverse platforms (e.g., Marketo, SFDC, Gainsight) into a single source of truth.

The mandate of data science is to unlock hidden patterns in data using powerful ML/AI algorithms. Larger and more integrated datasets increase the predictive power of these algorithms. RevOps & data science teams should partner on creating the most powerful datasets possible by using their combined organizational influence to access common data and align on standard definitions through shared data dictionaries.  

Coordinated Roles

Finally, RevOps/data science partnerships that have the most success occur when coordinated roles are established, and both sides participate in the important work of translating data insights to business needs. In terms of core responsibilities, RevOps has the primary business-facing responsibility. It is their job to understand what the business needs, engage their data science partners on where AI/ML can help, and translate the business need into technical requirements that a data science team can action.

To accomplish this, best-in-class RevOps teams are increasingly taking on more technical competencies to help in that translation work. Key competencies for the new RevOps org include:  

  1. Self-Service Automation & Analytics (e.g., tools like Alteryx or KNIME)  
  2. System admin/configuration capabilities (e.g., Ability to update and configure core systems such as CRM) 
  3. Business Intelligence (BI) & Data Warehouse Skills: (e.g., proficiency in BI tools like Tableau/Power BI and database languages like SQL) 
  4. IT Software Development Support (e.g., Access to IT/developer resources)  

Correspondingly, best-in-class data science teams do not wash their hands of the importance of understanding business context. Rather, they engage closely with their RevOps counterparts to understand the business needs and how the business leaders think/communicate. High-impact data science teams spend time in the field, go on ride-alongs, do customer interviews and experience data science solutions from the end users’ perspective to understand what truly creates value.  

 Data Science/RevOps Partnership in Action

Bringing together clear use cases, connected data and coordinated roles between RevOps and data science teams is challenging, but the returns are exponential. Recently, a major credit card provider experienced this success by doubling conversion rates with ML-optimized lead routing. Through this partnership, RevOps & data science jointly accomplished the following:  

  • Increased and integrated available prospect data  
  • Created business logic to align commercial motions by prospect quality  
  • Deployed those calls-to-action to sales resources who executed them  

Operationalize AI Models Through RevOps

Getting AI models off the shelf requires a deep partnership between data science and RevOps. Whether starting from scratch or optimizing a mature RevOps team, the following steps are essential:  

  1. Establish a RevOps blueprint 
  2. Align on the most pressing AI use cases  
  3. Build common data sets applicable to the identified use case 
  4. Foster partnerships between RevOps and data science to pull through value  

 

Need Help?

For more information about how Alexander Group can support the process of transforming data insight into realized business value, contact a Technology practice lead today. 

Related Insights:

Back to Top