UserT Testing strengthens ML-driven UX testing with $1.3 billion acquisition by Sunstone, Thoma Bravo

Gaining insights into how users experience and use the software is only possible by having humans perform all of the user testing. With the advent of modern sentiment analysis and machine learning (ML) techniques, ever more insight can be gained from testing.

User Check is one of the pioneers in the field of using ML techniques to help discover and analyze user behavior. The past two years have been a whirlwind for the company. In 2020, Test Users raise $100 million in funding, and a year later in 2021 the company listed on the New York Stock Exchange (NYSE) under the symbol USER.

Today, UserTesting announced that it has signed an agreement to be acquired for $1.3 billion by Thomas Bravo and Sunstone Partner. When the transaction is over, the plan is to merge UserZoom – which Thoma Bravo acquired in April 2022 – with UserT Testing, to create an even larger set of capabilities for user experience testing.

Andy MacMillan, CEO of UserTesting, told VentureBeat: “We’re in a space where we’ve built a suite of technologies to gather a type of feedback that we call experiential narratives. client. “UserZoom features a range of different complementary research techniques and methods that can complement some of our customer experience stories.”

How UserT Testing integrates ML

Over the past two years, UserTesting has invested significantly in technology that helps it distill insights from its testing.

The experiment involves recording users to see how they interact with apps, including what they click, and asking users to recount their experience. MacMillan says his company has invested in using ML to help extract insights into recorded user experience content.

“We are actually using unstructured content, but turn it into something structured,” says MacMillan. “We trained a set of machine learning models to help uncover what we call moments of insight.”

Moments of insight are sequences of information that can help identify trends that will improve user experience. Test Users using a variety of ML technologies, including natural language processing (NLP), computer vision and intent and behavior analysis.

Among the things ML enables UserT Testing is the ability to perform click path analysis, being able to track where the user goes and what they are actually trying to do when they click something. Analyzing user emotions is another key attribute that ML supports, as well as the ability to see if users are satisfied with the experience.

Taking it a step further, UserT Testing uses ML to help create visualizations that overlay intent and path behavior to gain insight into how people go through a website or an app.

“There’s a lot that we can identify about the behaviors we’re seeing people exhibit, while they’re going through a process,” he said.

ML’s Virtue Cycle

ML does not exist in a vacuum; by definition it’s about machine learning from data.

MacMillan explains that User Testing for ML is an ethical cycle where the models his company builds are continuously validated and extended with new data from human testing sessions. users have benefited from ML. He added that the ability for humans to validate ML models with their own eyes helps build trust in the models.

“We collect these customer experience scenarios – the kind of end-to-end video – and we use machine learning models to direct people to moments of greater understanding,” says MacMillan. “But you can always dig deep, you can always say ‘oh the model said, let me take a look at this customer experience narrative’ and see if the intent really aligns with the emotion. or not.”

In MacMillan’s view, one of the biggest challenges with ML for any organization is having the right kind of training data. Test users already have video recording, show what’s happening on the screen, and the test also collects click data from users. The tests are done based on the test plan, so there is a basic expectation for what the user has to do. UserT Testing has a dedicated staff that also labels content as part of their daily work to help train and optimize models.

“The purpose of the product is to help connect live teams with real customers and real people to get human insight into the product,” says MacMillan. “We think machine learning is really just a means of helping people connect with those moments of insight, but those moments are still human.”