Enterprise SaaS · Compliance · Elsevier · 2025 to 2026
Pure: where the data lives
Customers were asking where Pure's servers were and how it was hosted. The answers were not all written down, so I went and found them, and checked them against Korean data handling policy before rollout.
- Role
- Product Specialist
- Org
- Elsevier
- Platform
- Pure · AWS hosted B2B SaaS
- Worked with
- Teams in the UK, Denmark and Singapore
The question
Pure is Elsevier's research information management platform, hosted on AWS and used by research institutions worldwide. Customers came to support with questions that sound simple: where are the servers, and how is Pure hosted.
Support needs two things to answer them. A clear answer, and instructions that match how Pure is actually configured. Without both, a customer gets a vague reply or a set of steps that do not work on their setup.
The answers existed. They were spread across documents few people read closely and teams support did not usually talk to.
What I did
I started with the live documents in Pure's support hub, the ones written for customers and specialists. Read closely, they held detail that not every account manager or product manager knew.
Where the documents stopped, I went straight to the configuration teams and asked about the AWS server settings: where the servers sit and how Pure is hosted.
For the Korean market, I then verified Pure's configuration against South Korean government data handling policy for workflows that pull in external data sources, and worked through the documentation and compliance checks before rollout.
How the answer was put together
- AWS
- Amazon Web Services. The cloud provider Pure runs on. Where its servers physically sit decides which data rules apply.
- Configuration
- How a particular deployment of Pure is set up. Two customers on the same product can have different settings, so instructions have to match the one in front of you.
- Data handling policy
- Government rules on how data may be stored, moved and processed. South Korea's apply to workflows that bring in data from outside sources.
Why it mattered
Without it, support could not answer customers with any clarity, or give them instructions that lined up with the configuration. Technical support on those accounts would have been delayed while someone went and found out.