Understanding the investigative journey

Phase 01: Directory

Assessing access control

How do users view trust and validity in this context? What kind of sources do they want to see? What do they want to provide? Where is the line between organization and chainalysis control vs. crowdsourced information?

Initial Proposal: Linkedin style

The LinkedIn approach allows for the most optimal way a user would want to think about this data- getting the main “meat and potatoes” directly from the organization itself, verified, and then community crowdsourced data around legal process where applicable. This method does not come without technical undertaking.

A pivot towards simplicity

Too much complexity! While creating segmented access control may provide powerful mechanisms of information trust and validity, and foster future correspondence for public sector customers, as of today we have no backend capabilities to support role based permissions.

Assessing use cases

It takes a village. With four different products in production, evaluating use cases takes the input of several designers, PM’s and cross-functional stakeholders.

Optimizing for use cases

While creating a static directory inherently provides more value, how can we leverage the core use cases and predict user’s intentions by offering the right next step in their workflows? More often than not, investigators are looking at directory information in the context of a cluster on a graph. How can we give them insights into what clusters have directory data available?

Assessing component frameworks

End Result: Universal modeless dialog. Some of the design tenets established for this first cross-product initiative were:

• Directory will have a consistent design across all of the products in a step towards a more cohesive product suite.

• Directory will be accessible from any product, anywhere.

  1. As it is not the core use case but rather a supplement to a users workflow, modularizing this experience and integrating into our “service” views as opposed to being its own experience is the best approach. Metadata contextualized with on-chain data is more powerful (more on that below).

Crafting a beta: open-ended by design

he goal of launching a beta to a small subset of trusted customers was to start seeding the Directory with as much data as we can capture as possible. I began to think about what the update process means and what that represents in this learning phase of the product. Do we want it to be entirely frictionless? Do we want it to be as open-ended as possible to see what users tell us rather than providing a framework to get narrow results? How do we assess the tradeoffs of both?

Findings from open-eneded approach

Several trends started to emerge from the open-ended approach, which will help guide the way to crystalize the Directory back-end.

• Organizations updating their own org data

  • “We have a dedicated email address for compliance. the email is compliance@vauld.com.”

• Organizations confirming their own org data

  • “I am the Head of Financial Crimes Compliance and can confirm this is the email for subpoena submission.”

• Law enforcement updating external private sector org

  • “I am an investigator and this is what we have on record as the POC. Disclaimer: I personally haven't sent Whitebit an email.”

• Legal request information

  • "Former CSM + employee of Kraken. this is where they've asked to target compliance inquiries in the past. other link is to their institutions page which is not where compliance inquiries would go.”

Beta to GA

Directory: A final look

Phase 02: legal request

Lead with Research, not solutions

There was a lot of push to create an “auto-subpoena” that magically puts all customer data from Chainalysis products into a subpoena. It had allure because it sounded flashy and seemed like a sweeping way to solve a customer pain point. However, I wanted to make sure we were really solving for a problem customers had, and in the right way.

After concept testing several different methods, I learned that the subpoena process changes from country to country, and that different sized organizations have different feelings around the “auto” part of the process. The root pain point was being able to select the right info from Chainalysis products without having to select ALL info. Customers wanted an “add to cart as you go” method.

Just a piece of the puzzle

Ultimately, it made sense to hold off on the legal process workflow until we began to tackle case management. Directory is a strong foundational building block to the unified case management vision, and the legal process workflow will be a large part of that discovery work. However, it is important that we build the two in parallel.