Legal operations can get improved workflow design and buy-in from attorneys and other users without sacrificing speed by launching a process or tool and then iterating as data comes in, Daniel Tarle, senior legal ops analyst at workflow software company Asana, said in an Inside Voices podcast.
You might want to test a process or tool to death before launching to ensure it works for your attorneys or other stakeholders, like the sales team. But another way to do it is to collect just enough data to launch and then make improvements as real-time data comes in from actual users.
“It might seem that trying to build thoughtful, scalable processes are at the sacrifice of velocity, but why should that be?” said Tarle, who joined Asana two years ago from a legal ops role at Block, the fintech.
Since Asana itself provides a workflow software, Tarle has been integrating Ironclad contract lifecycle management (CLM) software with it and also with Salesforce and business-spend management software Coupa.
With the integration, the sales team has a seamless, self-serve workflow from Salesforce to Ironclad.
“It’s a full end-to-end process that can start in Salesforce, Asana or Ironclad,” he said. “Although there is value to having a single point of entry, this was intentional in terms of creating flexibility, depending on the type of deal or the different type of team that might be submitting the request.”
Legal ops pros can be at risk of analysis paralysis if they delay launch until they have everything worked out based on data generated from sandbox testing, user acceptance testing (UAT) or another type of testing.
What’s better, at least for the way Tarle likes to work, is to launch sooner rather than later and then iterate. That way the data you’re basing your changes on is from actual users trying to navigate the system.
“When people are really doing something that’s part of their core work, that’s where you can get data and feedback that helps keep everything moving and hitting that velocity,” he said.
At the same time, you need to have a robust human component, no matter how self-serve the process is, so users can feel comfortable enough to spend time getting familiar with the new workflow.
“You want to make sure folks know who you are, who built this, who’s maintaining it and who they can go to if they have questions,” Tarle said. “It’s important for them to know we built this not just for our needs but for their needs. We want to provide them with what they need to get their job done.”
Much of the onboarding process is self-serve, with templates and workflows that people use, but both formalized in-person training and informal question-answering over Slack and other tools are also a key component because that’s where you can build trust with users that they have support available to them.
“We build out [training] materials … and also provide periodic live trainings, where we go through more details, Q&A type stuff,” Tarle said.
To make high-velocity launches work, though, you have to be prepared to watch the data closely as it’s generated so you have a good picture of what changes are needed.
“We know a lot of the learning will come from actually doing,” Tarle said. That means “getting just enough data to launch something and enough to get going with your onboarding, then jump into the deep end. You have to believe in getting your hands dirty – jumping in and doing the work. That’s one of the best ways to accelerate that learning curve.”
“That's part of what success looks like to me,” he added. That’s “being able to see it functioning but also having the capability and agility to continually improve on it.”