BetterCloud Secure: Redesigning Risk Remediation for IT Teams

BetterCloud is a SaaS management product that helps IT teams automate manual tasks and enforce company security policies. Its Secure product could show an admin exposed files, misconfigured apps, and policy violations. It could not fix them. Remediation meant leaving the product, switching tools, and coming back to see whether anything had changed. BetterCloud was redesigning Secure to close that gap, from a passive monitor into an active guardian of sensitive data, and I was the Lead Product Designer, leading the design process and defining the remediation journey.

Starting with people

BetterCloud already held hundreds of user interviews and support calls. Raw, they were a pile. The Business Intelligence team and I worked through them together until patterns showed in what admins wanted, where they got stuck, and what frustrated them. From those patterns I built actor sheets with goals, quotes, and an imagined workspace for each person.

View larger (opens the full image)

Later decisions were checked against those sheets, which kept the design tied to real admins and not a hypothetical user.

Counting what admins faced

With the actor sheets in hand, I walked through the existing experience the way an admin would. Old Secure notified admins of risky documents and where they could be found. Remediation required going to the relevant source application. The menu pictured below is a recreated depiction whose exact product/version context is unknown; it does not establish remediation inside old Secure.

I audited the remediation action set by hand: 86 possible actions across five applications. That count describes the complexity admins faced across source applications.

That count made the challenge concrete: simplify the choices without losing the coverage IT admins actually needed.

One room, one goal

I facilitated a design sprint with designers, product managers, and customer-facing roles together. The goal on the whiteboard was to guide users to quickly remove 100% of risks. The questions beside it were honest ones. Is the right action the same for every user? How could bulk actions stay digestible? Is inbox zero even realistic? Story mapping and rapid sketching brought each person’s perspective into the same set of ideas.

Three capabilities came out of it. Quick Actions let a user take the correct action on a risk in as few clicks as possible. Risk Lifecycle Tabs showed current risks and successfully remediated ones side by side. Universal Action Properties gave every asset type and application one consistent way to configure an action, to lower the load on the admin.

The tabs rest on one rule about an asset’s life. It is an active risk until an action completes and it no longer violates the policy. An action can be in progress or in error along the way, and those are still active risks. Only the last step makes it secured.

View larger (opens the full image)

From storyboard to clickable prototype

I led the move from storyboards to a clickable Figma prototype. Each designer took ownership of a section, and I kept the full journey consistent from one section to the next. Users were already scheduled to review it, so the pace was tight.

Then the sprint’s bold vision had to meet reality. I worked with the Product Manager and the architects to map technical feasibility, and the work was cut into chunks that later became development epics. The aim was to keep the high-value capabilities without overextending engineering.

What changed between builds

The tab model did not stay put. The sprint prototype had five tabs: Active, Investigating, Secured, Allowed, and Activity Log. A later design has four: Recent Risks, Total Risks, Secured, and Activity. The insight cards gained counts of the sensitive content found, such as Social Security numbers and credit cards. I can’t tie each change to a single test session, so I show what changed and don’t claim to know why.

The risk inbox

The risk inbox is where I focused the data display and usability work. The tabs carry counts. The filters narrow a list that can run to tens of thousands of files. Bulk selection sits directly above rows sorted by severity, and each row shows the owner, who the file is shared with, when it was last modified, and the sensitive content inside it. Allison Kim designed the data visualizations, the charts. The list, the actions on it, and the usability are mine.

1234

1. 1Lifecycle tabs, each with a count: current risks, all risks, secured risks, and activity are one click apart, so an admin can see what is open and what has been dealt with.
2. 2Filters: Action status, Content Violations, owners, collaborators, departments, and asset types. A risk whose action ended in an error stays in the active list and is found by filtering on action status.
3. 3Bulk selection sits directly above the list: select assets, then act on all of them. Choosing the right action for 50+ selected files was the task whose accuracy I tracked.
4. 4One row per file, sorted by severity: name, owner, who it is shared with, last modified, and the sensitive content found, such as credit cards and SSNs.

Acting on a risk takes two steps. An admin selects from the list, then reviews and configures the action in a panel that sets the new sharing level and links to the API calls the action will make, before anyone takes it.

Putting each build in front of admins

From April to October 2022, I ran four design sprints, and engineering coded seven alpha versions of the interface from what the sprints produced. Each build was tested before the next. Five external customer-partner companies took part, two to three system admins from each, working through specific tasks: finding files, bulk-securing files, setting notifications, and walking through their existing remediation process outside the tool. Internal professional services staff, who support customer setup, tested alongside them. Between sprints, two to three IT admin users sat with the team each week to react to the Figma prototype.

The three capabilities held up. Admins got a comprehensive view of current risks, action properties that worked the same across applications, and a lifecycle view for tracking progress and auditing their actions over time.

Two numbers came out of the testing, tracked in aggregate by the Product Manager. New users averaged approximately 3.8 seconds to complete the tested task, missing the 3-second target; returning users completed it in under a second. The seven Figma/sprint participants and ten alpha participants overlapped, so this was not an independent 17-person sample or a controlled study. The measured task was not the complete bulk-remediation workflow. Detailed task definitions and timing/start-stop instrumentation are unavailable.

Task accuracy showed a clearer trend. Early tasks like selecting 50+ files and choosing the correct bulk remediation action had a 68% success rate, with most failures coming from misclicks. That climbed to 82% by alpha build 7, across 13 users tested between April and October 2022.

What didn’t hold, and what got built

The redesign tested well and drew real interest from the partner companies. Then budget constraints and a strategic shift in the company’s market focus shelved the project indefinitely. Several pieces did reach beta: the new secure file grid with inline actions, bulk actions, tabs that grouped secure file categories, and the first file-count insights. The Secure home with deeper insights was handed off to engineering but never built. Nothing reached a full production release.

	Recent Risks	Total Risks	Secured
Data Shared Publicly	View larger (opens the full image) Recent Risks, recreated	View larger (opens the full image) Total Risks, original screenshot	View larger (opens the full image) Secured, recreated

Not everything held up under testing. Data-refresh performance never met the mark. The target was under 50 seconds for infinite-scroll file loading, the same number users independently named as their own expectation. The best the team reached was on-request-only refreshes averaging two to four minutes to download, cache, and rehydrate. The cause sat in the backend, in caching file data so it stayed current and in sync with customers’ own systems, not in the design, and it held up the full release.

Universal Action Properties was meant to collapse the 86 audited actions into a much smaller set of reusable interaction patterns. User interviews suggested roughly a 70% reduction in the distinct actions someone would need to learn. That is a projection from interviews. It was never measured in production, because the full set never reached production.

What outlasted the product

The most tangible legacy was Fulcrum, the Figma design system built to support the work, with its Gatsby-based Fulcrum Guide. My recollection that high-fidelity prototypes went from weeks to days is qualitative, not an instrumented or universal measure. Separately, I estimate teams reached testable prototypes about 40% faster across three Q3 2022 sprints, two Secure and one Workflow Management; those sprints were not timed. The estimate does not mathematically explain the weeks-to-days recollection.

Fulcrum served the design team. Jason R. helped me build coded components as a proof of concept that never entered a product codebase. Organized engineering adoption did not happen, although individual developers informally reused Fulcrum-aligned components and color tokens. Former colleagues reported continued use after I left BetterCloud in March 2023; I do not have firsthand knowledge of later Secure implementations.