Increasing Efficiency Using Amazon OpenSearch Service with Aro

Aro’s broker services—which it offers both directly to customers and through business partners—uses multiple datasets to accurately match customers to financial products and credit opportunities. Customers input data such as their income and what kinds of products they are looking for—for example, loans, credit cards, and car financing. Aro’s solution then matches customer eligibility with lender criteria to around 150 products from more than 50 lenders to show customers the financial solutions that best suit their needs. The company delivers personal credit solutions 24/7 for the 40 partners and over a million customers who use the service.

Aro wants to offer the most comprehensive search results for its customers. “It's our mission to deliver personal credit solutions 24/7, and this can only be done by having the broadest and deepest available products so that, when customers in the United Kingdom come to us, they know that we’re searching everything,” says Nick Allen, Chief Technology Officer at Aro.

To accomplish this mission with its small team—around 150 employees—the company must maximize its efficiency. It does this, in part, by using AWS. Aro has used AWS services since 2015, and in 2018 it migrated its production workloads to AWS, offloading much of its infrastructure maintenance to AWS managed services. “Wherever we can use AWS services, we benefit from not having an ongoing maintenance effort for individuals within the team,” says Allen.

In 2022, Aro decided to migrate the search function of its analytics landscape from self-managed clusters on Elasticsearch to managed clusters in Amazon OpenSearch Service. The service could provide the same search capabilities as its previous solution, but with reduced costs and effort for Aro.

Aro completed the migration to Amazon OpenSearch Service in 3 months. The company was collecting data from 50 different applications in its cluster. To minimize interruption and prepare for the migration, Aro replicated its data from its previous solution into Amazon OpenSearch Service. “We took our time so that, when we cut over, it was seamless for our users,” says Allen. Modernizing its infrastructure using Amazon OpenSearch Service also sped up the time to onboard new embedded finance partners from weeks to days.

To accurately match customers with the best financial products for their needs, Aro collects half a terabyte of data into a single source of truth using Amazon Redshift, a cloud data warehousing solution that uses SQL to analyze structured and semistructured data. “We’re able to curate which lenders and products are available for any particular partner,” says Allen. “We selected Amazon Redshift for the power it gives us to process large amounts of information quickly.” In fact, some analysis jobs that used to take 8–9 hours for Aro to run now take under 1 hour to run in Amazon Redshift.

Aro queries the data using Amazon OpenSearch Service which has made their operations cheaper and easier. “Using Amazon OpenSearch Service, we can capture the activity from all our microservices and bring it into one place, which makes it simple for our engineering team to understand what is going on,” says Allen.
“AWS has provided us with all the building blocks we need to create the services that our customers use,” says Allen. “Our services that are used today in our digital context would exist without AWS.”

Using AWS infrastructure and managed services, Aro has significantly reduced the time that its personnel spend on maintenance and upgrade tasks. In 2017, a six- or seven-person infrastructure team spent 80 percent of its time on patching, upgrading, and maintenance. Now, in 2023, a team of two people spends 40 percent of its time on these kinds of tasks. With so much time saved, employees can instead focus on innovating for customers, adding value to the company rather than just keeping things running. “We’re able to use AWS to minimize the cost in terms of people’s time, as well as the capital expenditure for services that we would have to acquire or build without AWS,” says Allen.

Aro also uses AWS to maintain high availability—over 99.9 percent uptime—by deploying in multiple Availability Zones. It uses AWS best practices and component services to minimize security risks and build monitoring and security tools. “We get a lot of component parts from AWS that we can put together so that we don’t have to run a special unit dedicated to monitoring security,” says Allen. “This helps us maintain quality without additional time or resource.”