A performance review draft with a cross-functional reach signal connected to email header fields — subject, from, to, date — while message bodies are shown crossed out as never read.

Show cross-functional reach in reviews — from email headers, never contents

Q: Does PerfCopilot read my emails?

No. PerfCopilot reads email headers only — subject, from, to, and date. It never opens, reads, stores, or quotes the body of any message, and it never touches attachments. We read email headers, never the contents.

Q: Can email alone justify a performance rating?

No. Email reach is a supporting signal only. Communication volume is easy to over-weight, so the bias check explicitly guards against it, and the draft is built mostly from real work output. Email corroborates reach; it never carries the rating by itself.

PerfCopilot reads only your team's email headers — subject lines, who emailed whom, and when — to surface how broadly and how consistently someone coordinates across the org. It never opens a single message body. Then it folds that reach signal into a bias-checked draft built mostly from real work output.

Connect email — free up to 5 seats → · read-only, headers only, no credit card

✓ Headers only, never contents · ✓ Bias-checked · ✓ Combines with work output

What PerfCopilot reads from email (and what it never reads)

This is the part most "email analytics" tools won't put in writing. Here it is, plainly.

What PerfCopilot reads (headers only):

Subject line — the topic of a thread, for grouping coordination by theme
From — who initiated
To / Cc — who was coordinated with, and how wide the circle was
Date / timestamp — when, and how consistently over the cycle

What PerfCopilot never reads:

Email bodies — the actual text of any message
Attachments — files, documents, images
Message contents of any kind — nothing inside the email is read, stored, or quoted

Because we never read the body, PerfCopilot cannot and does not claim to cite what was written in an email. Email contributes a reach and cadence pattern — who someone works across and how steadily — not a quote. That honesty is the whole point: it is the only email signal you can defend in a calibration room without a privacy fight. Want the full source list? See all integrations.

Why email is a reach signal, not the review

Work output tells you what someone shipped. Email headers tell you who they had to pull together to ship it. A staff engineer whose headers show steady coordination with Product, Support, and two partner teams is doing cross-functional glue work that pure code or ticket data misses entirely — and that's exactly the contribution that goes unwritten and unrewarded.

But reach is a supporting signal, never a standalone rating. Volume is easy to game and easy to over-weight: more emails is not more impact, and a quiet specialist is not a low performer. So PerfCopilot's bias check explicitly guards against letting communication volume inflate (or deflate) a rating, and the draft leans on actual work-output evidence — GitHub, Jira, docs — with email reach as corroborating context, not the headline.

The combination is where it earns its place: email reach next to cited work artifacts shows both what shipped and how broadly it was coordinated — paired with the other 17 integrations for the full picture.

Generic email analytics vs. a bias-checked review draft

	Generic email analytics tools	PerfCopilot
Reads email activity	✓	✓
Privacy scope	often scans full message contents	headers only — subject, from, to, date; never bodies or attachments
Bias & fairness check (guards against over-weighting volume)	✗	✓
Combines email reach with work-output evidence for a full review	✗	✓ (18 integrations)
Writes the review (not just a dashboard of email stats)	✗	✓
No migration — exports into Lattice / 15Five / BambooHR	✗	✓
Cost	varies, often per-seat enterprise	Free up to 5 seats · Pro $4.99/seat/mo billed annually

PerfCopilot maps a review cycle of cross-functional reach into a cited, bias-checked draft in under 5 minutes — from email headers alone, no inbox reading.

Why "headers only" matters

The privacy guarantee is structural, not a setting. Because PerfCopilot only ever requests header metadata, message contents are never accessed in the first place. Plainly: we read email headers, never the contents. That is the difference between a fair reach signal and surveillance of your inbox.

Run the performance review generator across email reach and every work-output source at once, then see pricing — free up to 5 seats.

Connect email — free up to 5 seats → · headers only

FAQ

Does PerfCopilot read my emails? No. PerfCopilot reads email headers only — subject, from, to, and date. It never opens, reads, stores, or quotes the body of any message, and it never touches attachments. We read email headers, never the contents.

Then what does it actually read from email? Four header fields: the subject line, who the message was from, who it was to or cc'd, and the timestamp. From those it builds a picture of who someone coordinates with and how consistently — a reach and cadence signal, not the content of any conversation.

Can email alone justify a performance rating? No. Email reach is a supporting signal only. Communication volume is easy to over-weight, so the bias check explicitly guards against it, and the draft is built mostly from real work output. Email corroborates reach; it never carries the rating by itself.

Is this private and secure? Yes — and the privacy guarantee is structural, not a setting. Because PerfCopilot only ever requests header metadata, message contents are never accessed in the first place. See the /security page for the exact scopes and data handling.

Do I have to leave Lattice or 15Five? No. PerfCopilot drafts the review and exports it into the HR platform you already run — Lattice, 15Five, BambooHR, and more. Nothing to migrate.