Product quality is the heart of every company. Yet it has been an ambiguous characteristic to measure for many product and engineering teams. unitQs’ founders understood first hand how product quality can make or break a company from their successful nine year tenure at a mobile social network company, Skout. In their efforts to quantify product quality, they created the world’s first quality metric, the unitQ Score. The proprietary metric is a key performance indicator that represents the current state of a product’s health. It’s designed to equip businesses to take a data driven approach to their product quality efforts, act fast on user issues before they become widespread, and leverage indisputable data to pave fruitful avenues towards their product’s growth.
Technology Behind the unitQ Score and its Core Focus on Product Health
The unitQ Score is built from cutting edge NLP techniques and AI models that are constantly monitoring a wide-range of online feedback sources. One of its core capabilities include the ability to assimilate data that’s interspersed across channels, regions, and languages. Unlike any other metrics like star ratings or classifications, the unitQ Score is derived from data that meticulously focuses on reported online feedback related to product quality. It’s then able to provide prompt and imminent signals that businesses can utilize to drive quick action in triaging user issues and validating key product related decisions.
unitQ Score vs. Star Rating
Star classification is one way to capture a weighted average of how respondents rank businesses, typically from a one to five star scale, but it can take weeks for this rating system to aggregate feedback without prioritization of different reports. There’s also room for misunderstood information, mistranslations, skewed data, and misinterpretation of the ratings. Moreover, gamification aspects can come into play. They can be an amalgam of influences and incentives ranging from popularity, news and media, gift card or financial rewards. At times, there might not be any tangible rhyme or reason why a star rating has gone up or down. Consequently, businesses don’t get a concrete assessment or valuable signals that pertains to their product’s health.
The unitQ Score equips businesses with the biggest lens on all their user feedback across popular channels and 100+ languages. Teams can expect a single source of truth and actionable signals that’s directly correlated with product quality to quickly resolve user issues and foster product growth. Below are two demonstrations using current examples of how the unitQ Score can provide the most precise findings on a product’s health.
- We looked at an app that recently had a total of 1,783 reviews with a 4.5 average star rating in the iOS store. Although the overall rating was close to 5 stars (the highest score), which universally would indicate that a brand and product are doing great, there might be missed opportunities and signals to drive product growth without the unitQ Score Model. In the graph below, you’ll see that our AI performed an analysis on the 1,783 iOS reviews. Our NLP techniques are able to pick up on the context of languages and the general sentiment of every review with precision. Simultaneously, our machine learning models can filter out the noise and only focus on feedback related to product quality. As it analyzed each of those 1,783 reviews, the unitQ Score model found 51.5% of those reviews had actionable product quality issue signals (even with the higher star reviews), 21.4% had poor company sentiment, and 27.2% had no actionable description. A comprehensive report like this can help businesses fully capture and understand everything their users are saying about their product quality then leverage those insights to unlock their product’s fullest potential.
- Our team also examined recent data comparing the star rating and unitQ Score by looking at a free app in the Google Play Store. Earlier this quarter, Company A was tracking a low 2 star rating, while their unitQ Score ranked a high 97. Looking at their reviews on the Google Play Store would reveal their 2 star rating was influenced by an influx of recent feedback based on the company’s recent actions. While those reviews were relevant feedback, it didn’t represent anything about the product’s current health. The unitQ Score Model looked across all of the app’s recent product related feedback and determined everything was in very good standing when it came to its product quality.
Customer Taking Quick Action with unitQ Score
Recently, one of our customers had a severe drop in their unitQ Score. Our machine learning models were able to quickly pick up a spike in user data and feedback that helped them indicate an outage with their product. From there, their product and engineer teams were able to take swift action and fix the issue at hand.
In the graph below, you’ll also see a comparison between the unitQ Score and a star rating during the outage. As it was occurring, the star rating remained unchanged. Meanwhile, the unitQ Score provided an actionable signal early in the outage.
How to Check Your unitQ Score
Product quality matters more than ever. And now there’s an effective way to measure it. We encourage businesses to use the unitQ score to help drill down on your product’s health. We offer unitQ scorecards and update them monthly so you can benchmark your product quality efforts over time. Check your unitQ Score today to see how you compare to others within your industry.