PI Hire — Interview Builder
Overview
The Interview Builder uses a candidate's behavioral data to build job-specific interview scripts that get at the core behavioral attributes they want to see in a hire and understand how a person might perform under the specific circumstances of a given role.
The Problem(s):
- Inconsistent process - little to no best practices seen in interviews at most organizations.
- Lack of time and resources to plan for and conduct better interviews.
- Failure to convey interview learnings across hiring team effectively.
The Solution:
A robust tool for consistent, high-quality, risk-managed interviews. Interview scripts could be customized for each candidate in consideration for a given role. This helps train less experienced interviewers, mitigate risks (legal and candidate good will) from improper or irrelevant questions, and saves time for hiring managers, recruiters and other team members.
Learn more about the Interview Builder @ Predictive Index.
Humble Beginnings
This product concept came initially from UX research I did to support the HIRE product line, and quickly gained internal interest and priority.
Awful questions reported by dismayed HR professionals:
- Why should we hire you?
- Where do you see yourself in five years?
- What were you doing during this resume gap?
- How old are you?
- Do you have kids? (or plan to)?
- What’s your greatest weakness?
I partnered with product management, engineering and our internal human resources SMEs to pitch this feature, which was born from a light-bulb moment and some sketches based on feedback coming from HR professionals and hiring managers I'd interviewed in prior UX research work.
It proved to be a favorite feature amongst customers and had tons of headroom in terms of potential for future product and feature growth.
UX Research Informing the Interview Builder
I built out a research protocol, lined up observers, narrowed down our recruiting criteria and set up resarch interviews with participants who aligned closely with our target personas, but who had no direct experience with Predictive Index products.
I built out a research protocol, lined up observers, narrowed down our recruiting criteria and set up resarch interviews with participants who aligned closely with our target personas, but who had no direct experience with Predictive Index products.
Using the research plan, I crafted wireframes and screens in Figma to use during the research interviews, and we began running sessoins. I facilitated, and along with 1-2 internal observers we took extensive notes and collected them in Mural for later analysis, synthesis and discussion.
Product Team Collaboration
Working with the product owner and engineering team, I guided us through feature ranking and LOE estimation, as well as reviewing the research synthesis artfacts with the team.
Ultimately, I delivered detailed specs for the initial release as well as for several more iterations for future prioritization, where new features would be introduced and existing features enhanced. Sitting with the product team, we shipped the initial version and monitored feedback carefully.