Analytics adoption
10%→38%
+28 percentage points
01 · Alva Labs · recruitment analytics
Helping recruiters find and act on hiring bottlenecks.
- Analytics adoption
- 10% → 38%
- Hiring efficiency
- Time to hire, time to fill and candidate wait time improved. Results varied by industry, company size and role.
Recruiters struggled to identify hiring bottlenecks and decide what to improve. I took on product design, research and PM responsibilities, turning scattered data into clear, actionable insights across pipelines and the organization. In about a month and a half, I worked with T-Rex’s four developers, Customer Success, and customers to test and ship the web feature.
02 · Why it mattered
What’s slowing down hiring, and what can I change?
Signals
- Sales feedback
- Low dashboard adoption
- Customer questions
Research
Interviews with recruiters and TA leaders
How teams evaluate hiring, use reports and decide what to change.
Shared hiring data
Across company sizes, industries and countries.
The problem statement
Recruiters and TA teams can't tell what's working in their hiring process, don't know what good looks like, and lack the evidence to drive improvements. If we help them hire faster, better, and prove their impact, they're less likely to churn and more likely to expand.
- Business outcome
Reduce customer churn
- Business outcome
Increase expansion within existing accounts
Product outcome
Increase the share of recruiting teams using evidence to improve their hiring process
Retrospective refinement · not an originally measured result- Starting pointI can see hiring is slow, but I can’t tell where the delay comes from
- I find out candidates are stuck only after they’ve waited too long
- I can’t trust these numbers enough to act on them
- I don’t know how to judge the quality of candidates moving through our pipeline
- I don’t know whether the people we hired perform well in the role
- I can’t make a convincing case for changing our hiring process
- I can’t show how my work has improved hiring
03 · Design decisions
Finding delays and helping recruiters respond.
Problem 1I can see hiring is slow, but I can’t tell where the delay comes from.
Time to hire and time to fill showed how long hiring took. I broke down “speed” into measures that could show where time was being lost.
My measurement sketch: exploring speed from the recruiter, pipeline and candidate perspectives. What I heard
Recurring interview patterns
- Recruiters + managersResponsiveness
- CandidatesResponse time
- MarketApplicant volume
Missed time-to-hire targets
The solution
Four measures I defined for Alva
- Candidate wait timeNew visibility
- Time in stage
- Pipeline duration
- Throughput time
Familiar references retainedTime to hire · Time to fill
What I prioritized
The MVP decision
Clear to usersFeasible to buildPrioritize for MVP
Throughput timeDeferred
Hard to explain and implementCandidate wait time
Time candidates wait for the hiring team to act, score, or move them forward.
Time in stage
Total time each candidate spends in a hiring stage.
Time across assessments
Time to complete each assessment, including scoring by the hiring team when needed.
Hiring times & recruiter workload
Time to hire and fill, alongside each recruiter’s jobs, candidates, and hires.
Throughput time
The average time from entering the pipeline to a decision. We left it out of the MVP because it was too difficult to build and to explain to new users.
Hiring funnel 
How candidates move through the pipeline, and where they drop off.
Analytics 
Drag the divider or use the arrow keys to compare.Scroll inside After to explore. Before stays fixed.
Problem 2I find out candidates are stuck only after they’ve waited too long.
Recruiters already had plenty of admin. I wanted delays to surface with a clear next step, without making them search Analytics.
What I heard
Recurring interview themes
- Administrative work
- Searching Analytics
- Interpreting the metrics
Issues stayed hidden until someone looked.
The solution
A card recruiters could act on
Each card described the issue and suggested a response, like a status update, for the recruiter to approve.
Why I chose this
I explored different card types. Describing the issue and suggesting a response gave recruiters more help than a warning alone.
Proposed card designs 
Send a status update 
Request interview availability 
Send an approved offer Bringing the cards onto the homepage
The homepage brought issues and suggested next steps into recruiters’ daily work, so they didn’t have to remember who was waiting.
See what needs attention. Issues and suggested next steps, without digging through the data.
Homepage 
Drag the divider or use the arrow keys to compare.
Problem 3I can’t trust these numbers enough to act on them.
Time to hire and time to fill mattered to recruiters, but they didn’t tell the whole story.
I went back to the raw data and redefined the metrics around how candidates actually moved. That meant treating the pipeline as a graph, with steps people could skip or return to, rather than a straight line from application to hire.
04 · What changed
Outcomes
Business impact
2large contracts
won
Better Analytics helped us win these deals against a competitor.
Hiring efficiency
After 6 months- Time to hire
- Time to fill
- Candidate wait time
All three improved for customers who opted into the new Analytics. Results varied by industry, company size and role.
Customer Success
Fewer calls.
Fewer analytics issues.
Customer Success reported a substantial drop in analytics-related calls and problems.
Qualitative feedback
My contribution
Design, research
and product management.
When our PM left at the start, I took on all three roles for 3–4 months. I drove the project, working closely with my squad.
I worked with developers almost daily to check event tracking and clarify requirements, using maps, recorded walkthroughs and detailed discussions to explain the changes.
05 · Next exploration
What comes next
End of case studyAlva Analytics