Privacy-enhancing 

technology

Business growth and risk mitigation for the new data economy

scroll

website mockup.jpg
blank monitor.png
ezgif.com-gif-maker (1).gif

Overview and role

I am the lead and sole product designer at QEDIT working with the VP Product and development team to create platform/s for various uses cases. 

This is an ongoing project which I have been involved in from 2019 until now. 

Problem

Creating a usable platform around a technical solution. We had an API but no physical product to interact with. In order to sell this solution and provide real value to the users, we needed to create a UI in which to interact with this technical solution and to build out the value that it provides. 

 

The technical solution employs cryptographic techniques that allow you to keep PII private while still comparing and gaining insights on sensitive business information.

 

The design problem was that we could deliver a list of results that compared datasets via CLI, but it's not much of a product. We needed to create a product that reduced the need to spend time organizing datasets, would allow you to collaborate/compare datasets with other organizations, determine the results of these comparisons, analyze the results, and automate processes depending on the outcome of these results. 

 

Additionally, because we had many different use cases, it was a design problem figuring out how to create several different platforms to fit each of their individual needs.

cli.png
Capture.PNG

Scale of awareness

Our potential users are here

Aware of problem but

not aware of a solution

Problem/Solution aware & using a competing product

Unaware of problem

Problem & solution aware but don't know or use a product for this solution

Design process: research

We had a technical solution without fully understanding who our potential customers were. We had target audiences that differed for the various use cases of the technology. The first step of my research process was to determine the ideal customer for each of these use cases.

 

Determined persona per use case

  • HR- certification management

  • Chief marketing officer- marketing/business growth

  • Risk and compliance teams- Insurance and financial services

 

Because privacy-enhancing technology is such a new area, there aren’t many products out there currently. Additionally, many organizations are aware of privacy regulations, but they aren’t aware of a possible solution yet. This is the awareness barometer that I use to measure what kind of questions to ask in potential customer interviews. 

Design process: how to use personas

Instead of defining static personas,I looked for the common ground between these use cases and what the people we built these personas around told me in order to define the base product. We conducted interviews with our ideal and potential customers in all of the use cases.

 

I built prototypes initially for each of the individual use cases from which we were able to understand that there was siginificant common ground between the goals of all of them. Enough commonality to build one core platform that could support multiple use cases.

 

From there we could test the prototype and find where these personas really veer and what their goals are. If a more customized solution is needed we would find that out through understanding how people interacted with the prototypes and later the MVP centered around a more unified and general product. 

Blacklist.png
Duplicate Claims – 5.png
AC- 15-01-20 – Microsoft view – 1.png

Initial sketches that led to individual prototypes that we built in order to understand the various use cases. Click to enlarge.

Video of the prototype I built to get internal buy-in on creating one plaform for all the use cases

IMG_20201027_141010.jpg
prototype_edited.jpg

Whiteboard to prototype of the MVP for the solution

POC Learnings

After showing the prototype and running a POC with two companies in the business growth use case, the main frustration was that the results didn't really reveal anything. They used identifiers (row IDs) that weren't directly associated to the IDs in the raw data so it required users needing to download and reincorporate these results into a platform like excel to work with and understand the value of the results. 

Click images below to enlarge:

QEDIT- Duplicate Invoices Results – 3.pn

Before

Results table is not usable in this state. The results must be downloaded then uploaded into an excel sheet in order to work with the data. 

The identifier (row_id) is not understood. Although it proves that the data is in fact encrypted, the end-user cannot easily decipher which person is associated with the row_id.

The matched criteria is not understood in this design. What is matched criteria?

QEDIT- Results Conditional Formatting.pn

After

Results can be filtered, sorted, highlighted, tagged, and arranged in order to facilitate usability and ease of visibility.

By implementing privacy-enhancing technologies, we are able to show real data without compromising any personally identifying information or data leaks through the system.

The matched criteria is completely visible, allowing the users to understand the risk of each case fully. 

Key problem solved: reduce time on task

ezgif.com-gif-maker.gif

Data Input:

Setting up data in order to collaborate takes a lot of time to makes sure all the datasets are formatted correctly. By allowing for integrations with common data sources and creating a platform that helps you easily check that the data inputs are formatted correctly, we can reduce the time on task. 

ezgif.com-gif-maker (2).gif

Automations:

Allow for users to be able to select from multiple different actions based on varying result outcomes.  Reducing time on task for the risk and compliance teams involved and allowing for integrations with platforms they already use internally to manage flagged or risky behavior. Dropdowns allow for them to select custom actions and New action button allows them to create new actions based on criteria of their choosing.

Outcome

The current secure data collaboration platform is being used in a MVP state by two companies in the business growth use case and undergoing pilots with the organizations in the insurance and financial services sector. 

 

Using tools like hotjar and google analytics, we were able to see that the dashboards were a place that users were spending more time and understood how we could build those out more. 

 

The POC phase is critical in gathering information about how to iterate on the next versions and make the product more fleshed out.

QEDIT- Dashboard.png

Dashboards

For the MVP we used a Quicksite integration for the dashboard in order to better determine what kind of graphs and charts would be more useful to users. Based on user feedback we know now that we can build more customized dashboard in future versions.

IMG_20210323_161726.jpg

Sketches and dashboard designs for future customizations

Verification – 2.png

Designing for emerging technologies

As a designer in a very innovative space and working with emerging technologies, my role as a designer is not just to create usable interfaces, but to create a solution users trust. Most users will not understand the underlying technologies at play, but they need to trust that their data is secure and that their goals are being achieved. 

In order to create trust, you need to play on defined methods in UX/UI to speak to users sense of familiarity and to be as transparent as possible. When you are able to explain to users what's happening in the background, they can trust the system more fully. 

My work at QEDIT doesn't end here.

I'd be happy to show you the Excel and Google Sheets add-in I created, the management console for private asset transfer on Blockchain, or the Hands on Lab demo I built for VMware.