This is a recap of a customer panel from our recent Unlock Innovation Forum. Watch it on demand here.
At the recent Unlock Innovation Forum, leaders from Thomson Reuters, Shaw Industries, and New York Life shared how they're using AI, more specifically Varicent’s ELT Assistant, to cut routine data work from their teams' plates. They also talked about what happened when they redirected that time toward strategic decisions.
Here’s what they told us.
When Documentation Keeps Pace With Changes, Teams Gain Confidence to Move Faster
For several teams on the panel, documentation was one of the first areas they used AI to tackle.
At New York Life, Wesley Loden, COO of Field Experience, described the challenge to keep documentation accurate while they modernized legacy systems, some more than 40 years old. As data pipelines expanded and logic evolved, documentation updates couldn't always keep up with system changes.
Written descriptions of data sources and calculations sometimes reflected an earlier version of the system, and when teams needed to make changes, they often couldn't rely on documentation alone to understand how the pipelines were configured.
With Varicent's ELT Assistant, the New York Life technology team began capturing documentation as part of the build itself. The system automatically records logic, sources, and transformations as pipelines change, rather than requiring manual reconstruction later. Over time, this made it easier for teams to understand what was in place before deciding to make additional changes.
“What’s been really interesting is how it’s helped with something that most people dread, and that is documentation,” Loden said. “The assistant has really standardized how we document, sped up the process, and created a much more consistent approach across all of our teams.”
Inherited Data Pipelines Don’t Have to Be a Black Box
At Shaw Industries, Sales Compensation Manager Daniel Errickson described inheriting pipelines that his team struggled to interpret. Without clear documentation, analysts spent hours trying to reverse-engineer what a pipeline was supposed to do and why it was built that way.
Instead of tracing each transformation manually, Errickson’s team began using ELT Assistant to analyze the logic inside inherited pipelines. By asking direct questions about how data moved through each step and what each transformation was doing, they could quickly see how the flow was structured and what it was built to accomplish.
“The assistant just makes that almost instantaneous,” he said. “You can really understand what the intention of this was and what it’s actually doing.”
When teams can quickly see what a pipeline is doing and why, analysts and operators can make changes without worrying about what might break or needing to track down the original developer.
Analysis Moves Faster When It Happens Inside the Workflow
For Thomson Reuters, the most immediate change was how quickly they could analyze data.
Brad Weber, Manager of Business Strategy and Operations, described how his team regularly works with complex data sets involving millions of rows. Using traditional methods, identifying errors or gaps requires analysts to spend hours sifting through large volumes of data for hours.
By pulling data directly into Varicent ELT and using the assistant to ask targeted questions, Weber’s team was able to analyze the data much faster.
“We’ve taken processes that would take an analyst four or five hours and cut them down to about 15 minutes, with really precise, accurate, and actionable results.”
When analysis dropped from hours to minutes, Weber’s team could handle ad hoc requests the same day instead of pushing them into a backlog.
Building Pipelines Becomes Less Manual Over Time
At New York Life, Varicent’s ELT Assistant also changed how the team built data pipelines and transformations in the first place.
Developers began using natural language to generate an initial draft of the pipeline, then refined and validated the logic as needed. Over time, this reduced manual effort the team needed to configure the data flow, test outputs, and deploy new pipelines.
“Our developers now can use natural language and just get a solid draft from the assistant, refine it, and validate it,” Wesley Loden, COO of Field Experience said. “We estimate this has led to around a 30% reduction in total development time.”
By reducing how much overhead teams faced when building and maintaining pipelines, developers could spend more time on decisions that moved the business forward.
How ELT Assistant Frees Teams to Focus on High-Value Work
All three panelists noted that these operational changes had a broader effect on how teams experienced their work.
New York Life's Wesley Loden described how removing low-value tasks changed collaboration and reduced stress. “It frees up the team to focus on business outcomes. You’re not falling behind. You’re not missing deadlines. You’re not adding unnecessary stress.”
Brad Weber echoed a similar shift at Thomson Reuters. “People actually enjoy going in and building these pipes. They want to solve complex problems now.”
Daniel Errickson, Sales Compensation Manager at Shaw Industries, pointed to a growing sense of momentum as the technology improved month over month.
“When you build something that frees up hours every month or every quarter, people want to be part of that. We’ve moved from uncertainty to excitement.”
Teams became more willing to tackle complex projects once they no longer faced friction in routine work.
Instead of spending days analyzing data gaps or reverse-engineering inherited pipelines, Weber's team at Thomson Reuters could close 100 cases at once by identifying patterns across the data. Errickson's team at Shaw Industries responded to tariff changes and built new incentives in days instead of weeks. And Loden's team at New York Life stayed ahead of schedule on a major succession calculator project by building pipelines from scratch with the assistant.
What These Teams Would Tell Others Considering ELT Assistant
We asked all three teams what they would share with other RevOps leaders considering Varicent’s ELT Assistant. They focused on how they approached the work once it became part of the process:
Start with small, clear questions
Teams found the ELT Assistant most effective when they treated it as part of the problem-solving process, not a one-shot solution. Breaking work into smaller steps and asking direct questions made results easier to validate and act on.
Bring it into the process earlier
Teams that used ELT Assistant at the start of pipeline work often discovered simpler approaches before complexity set in. They spent less time reworking decisions later because they tested assumptions earlier.
Expect iteration, not perfection
Teams learned how to work with the assistant over time. Small changes in how they phrased questions often unlocked better insights. At New York Life, shared prompt patterns helped build consistency across teams.
Use time saved deliberately
The biggest gains came when teams reinvested freed-up time into higher-value work. Faster documentation, analysis, and pipeline builds mattered most when they led to better decisions and collaboration.
Watch the Unlock Innovation Forum on demand to hear their full conversation and learn how teams are applying AI across sales planning, incentives, and forecasting to build more connected systems.