Yes. Firstweek reads your Slack, Salesforce and Confluence through read-only connections and writes each new hire one onboarding artifact: who owns what, why things are the way they are, and what's on fire now. Every claim links to its source. The hiring manager approves it before day one.

That's the short answer. The rest of this post covers what "pull it into one doc" means, how that differs from the tools you already have, and who checks the result.

What does it mean to "pull context into one doc"?

Your company already wrote everything down. Why the enterprise tier was repriced last spring is in a Slack thread. Who owns the largest accounts is in Salesforce. The reason the handoff process has nine steps is on a Confluence page with a title that made sense to someone in 2024.

Each tool holds a different part. Salesforce has what happened with each customer. Confluence has how things are supposed to work. Slack has why they actually work the way they do. A new hire needs all three, about their own corner of the company.

Nothing is missing. It's in three tools, in no particular order, and your new hire starts Monday.

Pulling context into one doc means reading all of it and writing down the parts one specific person needs:

  1. The company. What it sells, to whom, and what changed recently.
  2. The team. Who is on it, who owns what, and who to ask about which thing.
  3. The role. The accounts, projects or systems this person inherits, and the history behind each.
  4. The why. The decisions that explain how things work today.
  5. Right now. What's in flight, what's late, and what's on fire.

That is a writing job. It has always been a writing job. Managers were going to do it, right after they updated the wiki.

Only 12% of employees strongly agree their organization does a great job of onboarding new employees (Gallup). Most of those organizations have a checklist. The checklist is fine. It tells a new hire what to do. It can't tell them why any of it is the way it is.

Meanwhile the new hire is on payroll from day one, and until someone hands them that context, most of the salary buys questions. The onboarding cost calculator works out what that costs, using your own numbers.

A chatbot and a search box share one hidden requirement: you have to know the question. In week one, a new hire doesn't. They don't know the old discount policy was dropped for a reason, so they never ask why. They don't know which teammate quietly owns the enterprise accounts, so they ask the wrong person. Politely. Twice.

The tools themselves are good at their jobs. Glean answers questions across the whole company, with sources. Guru answers the questions your team already knows to ask, with citations. Keep them. They're for the day after a new hire knows what exists.

An onboarding artifact runs in the other order:

By the last page the new hire knows what exists, which is most of knowing what to ask. The longer comparisons are here: Firstweek vs. an AI chatbot and Firstweek vs. Glean.

What does it read, and what does it never read?

For the question in the title: Slack, Salesforce and Confluence. The connections page has the full list and what is read from each.

The rules are the same for every connection:

We read it, summarize it, and forget it. The raw messages never touch our database. What a security reviewer will want to know is on the security page, including what Firstweek doesn't have.

Who checks it before the new hire sees it?

The hiring manager. Every artifact, every time. Nothing reaches a new hire until their hiring manager approves it, and the manager can correct any statement before approving.

This step isn't decoration. A Slack thread from 2024 can describe a decision that was reversed in 2025, in a meeting, out loud. The manager knows that. The thread doesn't.

Citations make the check quick. Every claim links to the thread, page or record it came from, so the manager can click through, see who said it and when, and decide whether it still holds. The new hire can do the same later, which is how a document earns trust from someone who has been at the company for four days.

The manager approves it in ten minutes, instead of the three hours they were never going to spend writing it.

What does a finished artifact look like?

One document, for one named new hire, that reads as if a well-informed colleague spent a weekend writing it. Nobody's colleague has that weekend.

A typical claim looks like this (the example is invented):

The enterprise tier moved to annual-only billing in March, after two customers disputed mid-year changes. Source: the #pricing-decisions thread, March 12.

One sentence on what happened, a clause on why, and a link to where it was decided. Repeat that across the company, the team, the role and what's happening now, and a new hire has the story of what they're joining instead of a list of logins.

The sample artifact is a complete one, with its citations, written for a fictional company's new hire.

FAQ

Is Firstweek a chatbot?

No. Firstweek writes a finished onboarding artifact before day one, and nothing has to be typed to get it. Once a new hire has read it, they can ask follow-up questions grounded in the artifact.

Does Firstweek store our Slack messages?

No. Raw messages and documents are processed in memory and never stored. Only summaries persist, with citations linking each claim to its source.

Does it work for roles outside sales?

Yes. Each artifact is written for one new hire's role: AEs, CSMs, engineers, product managers, implementation leads and relationship managers. The sources in scope change with the role. The shape of the artifact doesn't.

How do we try it?

Run a pilot on one real hire who starts within 30 days. There is no card, and nothing converts automatically. At the end, you decide whether it's worth keeping.

Sources: Gallup, for the 12% figure.

Run it on your next hire. 30 days. Then we talk. Start a pilot or see a sample artifact. How it works for scaling teams.