Helping buyers evaluate a new AI category

BrightHire created interview intelligence. When buyers started searching for “AI interviewer,” it needed content that explained the shift, held up in a regulated space, and made its place in the market clear.

At a glance

Client
BrightHire by Zoom, interview intelligence
Work
Guides, comparisons, and alternatives pages across three content clusters
Engagement
2026
System
Wordful

Before

The turning point

Every claim needed a source, and contradictions went into a register instead of stopping the work. Research caught widely repeated statistics that didn't hold up before they reached a page, including a popular claim that people spot deepfakes about half the time.

Every competitor page also kept a section on where that competitor is strong. It made the comparisons fair, and both named-competitor pages cleared legal review.

What we built

After

Read the work

Read the full story

The problem

BrightHire coined “interview intelligence,” but the market was moving. Buyers had started searching for “AI interviewer,” and competitors were publishing for it. BrightHire needed to explain the new term, show where it fit, and help buyers compare options, all in a category where hiring regulations make loose claims a real risk.

The work ran asynchronously, with limited access to internal systems, so every piece had to start from what could be verified.

A register instead of a bottleneck

Every claim got a source and a verification state before it could be used. When the client's sources disagreed with each other, the conflict went into a register and the work kept moving, with a specific question for the right person.

Deep research on the candidate fraud cluster corrected or blocked at least one planned claim on every pass. Widely repeated figures that didn't survive a check never reached a page.

Fair comparisons that cleared legal

The comparison and alternatives pages named competitors, so each one went through competitive-truth and legal review. Every claim about a competitor came from that competitor's own public materials, and each page kept a section on where the competitor is strong. The legal gate, not the writing, set the pace. Both pages shipped.

A system that learned

Editor feedback became written rules in the system. From the first content cluster to the second, correction rounds per page fell by about two-thirds. The amount of editing per page stayed about the same; it arrived in one consolidated pass instead of several scattered ones.

What BrightHire keeps

  • A register of tracked claims, each traceable to every page that uses it
  • Templates for comparison and alternatives pages
  • Voice and style rules learned from its own editors

If this sounds like you

Your category is shifting, and every claim has to hold up? Tell us what you need. No prep needed.

Send a message

Prefer to talk? Book a free scoping call.

Tell us what you need

We'll only use your email to reply. Privacy policy