Email Insights

Stop guessing what's in your inbox history.
Measure it.

Ten years of email is too much to understand by scrolling. Email Insights turns preserved records into measurable patterns: who appears most, which organizations dominate different periods, how activity changes over time, where attachments cluster, and what kinds of messages actually make up the archive.

check_circle Measured archive facts check_circle People + organizations check_circle Trends over time
Most active relationships · 2022
Northwind
847
Sarah Jenkins
421
Archive composition
Conversation 42%
Bulk 31%
Notifications 18%
Other 9%

Questions a mailbox can answer when you treat it as history

The point is not to grade your productivity. It is to see patterns that are invisible message by message.

group

Who was actually present?

See prominent people, domains, and organizations across the archived period instead of relying on who happens to come to mind first.

show_chart

When did activity change?

Compare weeks, months, or longer periods to see spikes, quiet stretches, new relationships, fading relationships, and changing communication volume.

category

What is the archive made of?

Separate real correspondence, bulk mail, notifications, calendar messages, receipts, and other categories where current classification supports them.

attachment

Where are the artifacts?

See attachment counts and concentrations so periods or relationships with lots of documents become easier to investigate.

schedule

How did communication behave?

Where currently measured, response timing and thread activity can describe communication patterns without turning them into a judgment about performance or intent.

compare_arrows

How did one period differ from another?

Comparisons are where archive analytics become personal: this job versus the last one, this client relationship before and after a change, or one year against another.

Sometimes the question is bigger than one message

Search begins when you need one thing. Insights begin when you wonder what the whole record says.

work_history

"What did this job actually look like?"

Which people and organizations dominated? When were the intense periods? What changed over the years?

handshake

"Was this client always this important?"

Measure relationship activity across time before moving into a richer Organization Relationship History.

history

"I forgot how much of my life is in here."

For Collectors, the archive itself becomes interesting once years of people, places, work, purchases, and conversations are made visible.

Measured facts are the first intelligence layer

If the archive can calculate something directly, calculate it directly before asking a model to interpret it.

1

inventory_2 Preserve

Insights begin from records that were actually archived.

2

functions Measure

Aggregate people, organizations, dates, categories, attachments, thread and activity data using deterministic calculations where possible.

3

compare Compare

Make differences across periods or relationships visible instead of reducing the archive to one lifetime total.

4

auto_stories Interpret when useful

Use measured archive facts as source context for Calendar Insights Report and relationship reports rather than inventing a story first.

Email Insights FAQ

Are Email Insights AI-generated?expand_more
Many useful insights are deterministic measurements over archived records. Model-assisted interpretation should be a separate layer, not a requirement for basic counts, trends, people, organizations, dates, or categories.
Can Insights tell me why someone stopped replying?expand_more
Not as fact. AtArchive can show evidence such as declining message frequency, longer gaps, or changed meeting patterns where measured, but it should not invent motive or causality from those signals.
Can I compare multiple email accounts together?expand_more
Treat current analytics as archive/account scoped unless a combined cross-account view is explicitly verified. Multiple archives can still be explored separately and compared manually.