Setting the file. One moment.
Skill 29 · Knowledge Base For Startups
Subchapter 29.123
references/learn/genai/relative-insight-is-uncovering-the-why-behind-customer-feedback-with-generative-ai.mdMarkdown10 KBView on GitHub
How Relative Insight turns data into decisive action
Great businesses are great listeners. Paying attention and responding to customers, target audiences, and employees is essential to making a positive—and lasting—impression on those that engage with your organization. Surveys, online conversations, and customer service interactions all contain invaluable data that can help businesses create better experiences. However, turning unstructured qualitative data into actionable insights isn’t easy.
Together with Amazon Web Services (AWS) (opens in a new tab), the team at Relative Insight (opens in a new tab) is changing that. By leveraging high performance generative artificial intelligence (AI) (opens in a new tab) models in Amazon Bedrock (opens in a new tab), the company is transforming vast amounts of customer feedback into insightful narrative reports. With Relative Insight, businesses can deliver more valuable experiences and quickly become more responsive to the needs of the people that matter most—customers.
Founded in 2014 by CEO Ben Hookway, Relative Insight is a UK-based technology and analytics company that helps businesses uncover the “why” behind changing key performance indicators (KPIs). “Our mission is to elevate text analytics to improve businesses,” says Hookway. “That means two things: to improve the quality of text analytics, and to make sure it’s adopted as widely as possible adopted within companies.”
“Our key differentiator is our methodology,” says Hookway. “We compare language sets and then express the differences in the language as metrics. So, for example, your Net Promoter Score (NPS) surveys in May compared to your NPS surveys in April. The difference between the surveys is what you’re really interested in, and that’s what we produce.”
The company’s most recent offering, Accelerator AI (opens in a new tab), builds on top of its existing dashboard solution to increase the speed at which customers can uncover and act on insights hidden in unstructured text data. “Accelerator AI takes the metrics that we produce, combines them with raw text data, and automatically produces a narrative report,” says Hookway. “Those reports can then be distributed to stakeholders in the company.”
“Relative Insight has a number of customers across financial services, large pro sports teams, restaurants, and in the travel industry,” says Oliver Reihill, Head of Product. These businesses use Relative Insight to provide a more enjoyable experience for fans and customers. “Most pro sports teams do fan experience surveys at the end of every game. That data is fed into Relative Insight so that we can understand why the fan experience score is going up, going down, or staying the same,” says Hookway.
Imagine there’s a lot of discussion around the stadium’s public address (PA) system being too loud after a particular sports game—Relative Insight can be used to understand which section of the stadium is experiencing those issues, whether the individuals sat there are season ticket holders or not, how are old they are, and more. Accelerator AI uses that data to create detailed reports so the decision-makers can act quickly to improve fan experiences.
By comparing different metadata points, Relative Insight can also reveal what matters most to different customer segments. “If you do a comparison between season ticket holders and casual visitors, you will see a difference in what they care about,” says Hookway. “Casual visitors might complain more about transportation, getting to the car park, finding a parking space and so on. Whereas season ticket holders might be more concerned about the facilities, toilets, whether the line is too long, things like that.”
“As far as we understand, we’re the only text analytics platform that is doing these rapid but very detailed and actionable reports, and having that intelligence sent straight out to the people who can actually affect change,” says Hookway. “This is the key; there’s no point in having the analytics if you can’t affect the change.”
Developing any AI product requires access to high-performance models, trusted expertise, and secure, scalable infrastructure—Accelerator AI was no different. Relative Insight worked closely with AWS to bring its vision for the product to life. “AWS has been instrumental for Relative Insight,” says Hookway. “They have provided us with technical support while building on the AWS Cloud but also resources around how to innovate very quickly with generative AI, using Amazon Bedrock, and helping us with our go-to-market as well.”
The Relative Insight team took part in AWS Activate (opens in a new tab), a startup-launching program designed to help disruptors rapidly build, launch, and scale on AWS. Startups enrolled in the program can access AWS credits to help them get started with AWS services at no cost, and access proven expertise from AWS Support engineers. Relative Insight used AWS credits to fund experimentation with Amazon Bedrock.
“The AWS Activate program was really fantastic for us,” says Hookway. “Having had access to Amazon Bedrock, we turned a prototype into a product inside eight weeks.” Reihill adds: “That work involves a great deal of experimentation, working with new technology, trying new models, and AWS credits enabled us to do that in a cost effective and scalable way.”
The Relative Insight team is using Amazon Bedrock to access and experiment with high-performance foundation models (FMs), including large language models (LLMs) from leading providers. Amazon Bedrock is a fully managed service featuring a broad set of capabilities for building generative AI applications with security, privacy, and responsible AI. “Amazon Bedrock gives us the flexibility to pick the right model using AWS infrastructure while still maintaining the security and scalability that we need,” says Hookway.
“We were able to go to one service to develop against different large language models, as opposed to having to go to individual providers, which would mean more development time, as well as working through the various security and scalability issues,” says Reihill. “Not all large language models are created equal. For different tasks, you need different models. With Amazon Bedrock, we were able to experiment with models and find the right tool for the job. We saved time and effort and achieved cost savings of about ten percent.”
Beyond technology, Relative Insight has also recently joined the AWS Partner Network (APN) (opens in a new tab). By becoming an AWS Partner, Relative Insight has expanded its reach, delivered greater customer value, and driven profitable growth. “The AWS Partner Network has been fantastic for us for a couple of reasons,” says Hookway. “One is that it has facilitated introductions with the customers that we really want to talk to with that credibility of being an AWS Partner. It’s also introduced us to venture capital companies who are interested in funding Relative Insight as we scale up.”
Becoming an AWS Partner has also helped Relative Insight products to be made available on the AWS Marketplace (opens in a new tab). This has helped the company to increase its market exposure, build credibility, and ultimately win new business. “The AWS marketplace brings a lot of benefits for us,” says Hookway. “When you’re selling to enterprise vendors, they want to know that you’re secure, that you’re robust, and you’re scalable—the AWS Marketplace provides all of these things for us.”
The Relative Insight team were able to join the AWS Marketplace in a just a few weeks. “AWS is really an enabler,” says Reihill. “Getting us onto the marketplace is key. It allows us to fire up an instance very quickly for a large customer if they want to use our platform in their own instance of AWS. Put simply: it provides more ways for customers to buy Relative Insight products.”
Working with AWS has enabled Relative Insight to innovate, scale, and win new business. “If we weren’t on AWS, we really would have struggled because as we scaled up and got some really big customers, we would have been straining to support them with the amount of data that goes through Relative Insight,” says Hookway. “But with AWS, that’s been a seamless process. We can handle all that data with no problem.”
Moving forward, Relative Insight is continuing to collaborate with AWS as it improves its offerings and delivers more value to customers. “What’s next for Relative Insight is a real focus on eliminating all friction between the amazing insights we can generate and action in businesses. That means getting reports into the hands of people who need them and bringing the insights that we produce into other systems so that action is automatically taken,” says Hookway.
He continues: “AWS has really helped us with our go-to-market. It exposes us to a bigger breadth of buyers, and it’s given us a much stronger strategic outlook on what we can do and the benefits we can bring to enterprises.” Reihill adds: “The account management team and all of the solution architects have been invaluable, enabling us to do things that we couldn’t have done without AWS.”
“We’ve got an exciting roadmap heading into 2025. Accelerator AI is coming out now and we’re also looking at areas around voice and other ways of consuming data with our customers. So, it’s a very exciting year for taking our customers from data to knowledge,” says Reihill.
Author: AWS Editorial Team
The AWS Startups Content Marketing Team collaborates with startups of all sizes and across all sectors to deliver exceptional content that educates, entertains, and inspires.