Cancer Research uses AWS for payments and trial search function

0
101
Oracle enhances customer experience platform with a B2B refresh

Source is ComputerWeekly.com

Well-known charity Cancer Research has moved its donations platform on to AWS architecture, and has since been experimenting with ways to use the technology.

Cancer Research has been on a digital transformation journey for a long time – most organisations are in an ongoing state of transformation to match the changing needs and expectations of consumers.

Though Cancer Research isn’t a retailer, they are often held to the same standard by those looking to donate, making the online and mobile experiences extremely important.

In partnership with head of engineering Andrei Adler, head of platform engineering Pete Ainsworth developed a new engineering strategy for the charity, including a plan to upgrade the charity’s old payments management system.

Ainsworth says: “We inherited a set of products that had been built out around our core needs. So, our donation platform, online fundraising platform, our event management platform, covering key areas of the business for a charity organisation.

“Those had grown and evolved over time, we’d developed them quite rapidly from scratch, taking the functionality that used to be in our CRM and building them into independent products that could work independently and be developed on a much more rapid timeline.”

While these products served their purpose at the time they were created, the charity and its work began to grow too big and busy, making it difficult to maintain and scale the technology it has in place.

“It often would have downtimes,” Ainsworth says. “There was a lot of engineering effort for delivering less and less value as we were going along.”

But this meant a pivot in the charity’s initial strategy for technology – where before it has hoped to stay vendor neutral, it instead investigated what a cloud provider could offer.

“Avoiding that vendor lock in, at the time, made a lot of sense,” says Ainsworth. “But it led to us custom coding a lot of things that subsequently cloud service providers offered off the shelf as a service. We felt at that time that it was really an opportune moment to start using the native features of cloud service providers to our advantage.”

Improving the payments experience

One of Cancer Research’s main functions as a charity is to facilitate donations to fund the ongoing research into curing cancer – which involves handling consumer cash. Whenever making payments is involved, whether as a donation or in exchange for goods and services, processes need to be seamless and trustworthy or else consumers will be deterred from using the service.

Because of the nature of the technology left behind by previous projects, it made more sense for the team to completely rebuild the donations platform rather than re-engineer what already existed.

Cancer Research’s engineers were given some “high-level” guidelines about how the new donations platform should be developed – they chose to build the platform in AWS, and so where possible they were asked to use as many AWS features and “out-of-the-box” services as they could to avoid complexity.

They were also asked to develop it using JavaScript, one of the main skillsets of the engineering team, therefore making it easier in the long term for maintenance purposes, and to ensure the platform was cloud native and to make the platform serverless, giving as much of that responsibility to AWS as was possible to reduce maintenance costs.

When taking payments, firms must ensure their infrastructure is built in a certain way to meet legal PCI DSS compliances, but by using AWS, some of that responsibility was “offloaded”.

While it wasn’t the main aim of the engineering strategy, the charity has seen a cost saving from streamlining its use of technology in this way.

Ainsworth says: “It’s very fast, it’s very easy to update and – not necessarily as a as an original objective, but as a result – it’s ended up significantly cheaper to run. We’re looking at 94% cheaper than the previous infrastructure to run. As a kind of additional bonus, because it’s able to scale much better, it’s able to handle anything that we throw at it.”

Using telethons as an example, Ainsworth explains that regular Stand Up To Cancer events will cause large increases in usage of the donation platform, sometimes taking millions of transactions in one night, which are now much easier to handle with a scalable platform.

The process

So, how did Cancer Research go about moving from its own complex in-house systems to AWS’s cloud-based, serverless architecture?

“Initially, we started with a with a minimum viable product [MVP], which just included our single donation page, which is the main page that you land on – the front page of cancerresearchuk.org website,” Ainsworth says. “That single donation page offered a really straightforward, isolated case, to be able to move and to learn how to work in serverless, which is a new thing for us as an organisation.”

Once this page was performing satisfactorily on the new system, the engineering team started at add new instances – for example, recurring donations.

“We looked at the integrations with other platforms – for example, our online fundraising platform takes donations, and it uses the donation platform to do that,” says Ainsworth. “Gradually as we’re building out, it gets slightly more complex. But we’re learning more and more as we’re going along.

“What we found is that, at this point where the platform is relatively mature now, we’re able to move into new use cases extremely quickly. We’re able to adapt the platform very quickly to new use cases. Very recently, we’ve introduced direct debits, which was another one of our core use cases, and the delivery of that was exceptionally quick.”

Research trial finder

Based on what the engineering team created using AWS for its donation platform, Cancer Research then looked into how they could use the platform for other use cases across the charity.

Ainsworth says: “One area that has shown the real benefit of this architecture was in the Experimental Cancer Trial Finder project.”

Developed alongside the Experimental Cancer Medicine Centre, the Experimental Cancer Trial Finder is a platform that helps doctors find clinical trials for their patients.

Ainsworth explains that previously the process of finding a possible trial for a cancer patient was complex and heavy in paperwork, usually reliant on a doctor knowing that a particular trial was taking place that might benefit their patient specifically, and enrolling them in a trial was arduous.

“The basic premise of it is when oncologists are in surgery speaking to patients, the previous method of connecting them to clinical trials was very manual, it was shuffling paper around,” he says.

The platform currently has more than 600 trials, with research sites updating their profiles monthly to reflect the most up-to-date information about the trials taking place.

By digitising the process, more patients are now able to find trials that suit them. “It holds a database of really complex markers about the patient and has a database of clinical trials, so is able to match those up in a way that wouldn’t have been possible before,” says Ainsworth.

“It’s a really good example of how this architecture was able to really step us into innovative spaces that might have either taken us an awfully long time to get to a point of delivering value, or might not have happened at all in under previous architectures.”

As time has gone on and advancements in medicine have been made, Cancer Research has changed its tagline from “together we will beat cancer” to “together we are beating cancer”.

“It generated a bit of an emotional reaction that wasn’t expected when they launched it, I actually welled up. Working in cancer research, you’re very connected to the cause. But sometimes, you know, you get lost in the job and you forget why you’re doing it. And some something like that really reconnects you,” Ainsworth says.

Source is ComputerWeekly.com

Vorig artikelAzure Machine Learning – General Availability for November
Volgend artikelGeneral availability: Azure AI Advantage for Azure Cosmos DB