Who ends up in this room?

Research, insights, competitive-intelligence, enablement, and knowledge-management teams that work through large document sets like reports, transcripts, decks, and PDFs every week. Strategy teams fit well too, especially consultants and analysts building briefings from that material. We plan the group and a sample document set together so the labs match current research questions.

How is this different from your Google Workspace with Gemini training?

This program is scoped to bounded, source-grounded notebooks: a defined document set you question, cite, and turn into briefings or Audio Overviews. Google Workspace with Gemini training covers the broader productivity suite instead, Docs, Sheets, Slides, Gmail, Meet, and Drive. The two overlap on prompting and verification habits, but the tools and the workflows are different.

Does this assume any technical skill?

No. The training is for research and knowledge-work staff, and it assumes no coding, data-science, or engineering background. You only need a working knowledge of the documents and questions your team already deals with day to day.

Does this training cover building a custom RAG system?

No. This is a training program on the notebook workflow: sourcing, querying, citation checking, and outputs. If your team needs to build a custom RAG pipeline, that's separate engineering work outside this program's scope.

How do you handle confidential or client data during the training?

Exercises use approved non-sensitive or reduced examples, with synthetic document sets agreed with your team in advance. We don't process confidential or client data in the training environment. Personal data follows the same boundary. The only exception is when your organization has an approved environment and data policy that explicitly allows it.

Can the training be tailored to our research workflows?

It has to be, or the labs won't mean anything to your team. Before the session we run a discovery call to learn your document types and research questions. We also map the stakeholder outputs your team produces. Then we build the labs around a realistic version of that material instead of generic examples.

Does the program promise a certain accuracy or productivity gain?

We don't promise a specific productivity gain or guarantee accuracy or compliance. What we teach instead is the evaluation and citation-checking habit that keeps a human reviewer accountable for every claim a notebook produces. Zeo is not an official NotebookLM vendor partner, and Zeo is not a certified provider of any kind. The habit is the deliverable.

What do teams keep after the training?

A citation-checking checklist, a notebook-naming and organization template, a source-boundary worksheet for data and confidentiality decisions, and a pilot plan with a named owner and a measurement approach for the following quarter.

AI Training · NotebookLM

NotebookLM Training for Research & Knowledge Teams

Your team already spends hours cutting through reports, transcripts, and PDFs to find the one fact a stakeholder needs. NotebookLM organizes a defined source set into a notebook you can question, summarize, and use for briefings, but only if every answer stays traceable back to the source that supports it.

This program builds that discipline: sourcing a notebook correctly and checking citations before anyone repeats a claim. The workflow then becomes a team habit instead of one person's trick.

modules: 6
hours: 11

Some of the 500+ brands we've worked with

Selected brands we’ve worked with across industries.

See all references

Researchers want a fast way to confirm a fact without re-opening ten documents. Analysts want the same thing across a stack of transcripts and reports. Enablement and knowledge-management teams want everyone pulling from the same source set instead of five different personal notebooks. All three groups run into the same failure mode: NotebookLM answers fluently whether or not the source backs up the claim, and a well-written summary gets treated as fact before anyone opens the citation behind it. A briefing or Audio Overview built on an unchecked notebook can carry that same false confidence straight to a stakeholder who never sees the original document at all.

Pointed at a bounded set of sources, a language model summarizes, compares, and answers against material that doesn't change, turning hours of re-reading into a direct question. NotebookLM training turns that into a team habit: scope and source a notebook correctly, check every citation against its passage, and turn a verified notebook into briefings and Audio Overviews — without a custom RAG build, the broader Gemini suite, or open-web search. See Google Workspace with Gemini training for the wider stack.

The research budget goes to re-reading

Analysts lose hours re-opening the same reports and transcripts to confirm a number or trace a claim — a well-built notebook replaces that repeated search with one question asked of a bounded source set.

Fluent isn't the same as grounded

NotebookLM's answers read fluently whether or not the source backs them — opening every citation and checking the passage is the habit that matters before anyone repeats a claim.

Whose notebook is this, again?

Without a shared method, each person builds a one-off notebook with its own name, and the next researcher on the same topic just rebuilds it from scratch instead of reusing what already exists.

Boundaries blur fast

It's tempting to drop a client or personal file straight into the source set, even one never cleared for the notebook's audience — staff need a clear rule for what belongs, who can see it, and when to redact.

An Audio Overview needs the same fact-check as text

An Audio Overview or generated briefing can restate a source's claim more confidently than the source itself does — it needs the same citation-checking discipline as a written answer before it reaches a stakeholder.

01
NotebookLM foundations and boundaries
90 minBeginner
The module explains what a source-grounded notebook is and how it differs from general chat or open-web search. It also places NotebookLM beside your other AI tools. The room leaves with one clear sense of what NotebookLM is and isn't built to do.
What counts as a bounded, source-grounded notebook
NotebookLM versus a chat assistant or RAG
Questions it answers well
When a notebook is the wrong tool
A live walkthrough, start to finish
02
Setting up a notebook correctly
120 minBeginner
Setup decides whether a notebook is trustworthy. Your team chooses which documents belong and gives the notebook a name and description people can find. It also defines an approved source under your data rules. Participants build their first working notebook from a realistic document set.
A source set that actually answers the question
Approved formats: PDFs, transcripts, decks, web pages, audio
A name a team can find later
What client and confidential data can go in
Add sources, keep answers
A lab on a real document set
03
Querying sources with citations and grounding
150 minIntermediate
The core discipline of the workflow is asking a well-scoped question, then opening every citation to confirm the source says what the answer claims. Participants practice catching the gap between a fluent summary and what the source supports. Missing that gap is the single most common way a notebook workflow goes wrong.
A question scoped to what you need
Citations followed back to the source passage
When an answer overstates or blends its sources
Follow-up questions that narrow or stress-test an answer
One claim, checked across every source
A routine to run before any answer ships
04
Briefings and Audio Overviews for study guides
120 minIntermediate
Using a checked notebook for a stakeholder briefing or study guide, including an Audio Overview. Participants learn exactly what still needs a human edit before any of them go out, since generated formats carry the same grounding risk as a direct answer.
A briefing document, edited for accuracy
An Audio Overview, checked against its notebook
When a summary needs a full rewrite
Picking a briefing or study guide instead of audio
Disclosure, done appropriately
05
Shared notebooks and governance
90 minIntermediate
Moving from a personal notebook to a team resource means new rules: sharing, roles, ownership. The data and IP boundaries decide what a notebook can safely hold once more than one person can see it. None of that is optional once a notebook stops being one person's file.
A notebook shared, with enforced access limits
Who can add sources, who can edit
Security, confidentiality, IP
A scope note a new teammate can trust
Fixing a wrong answer in a shared notebook
06
Team rollout and a governed workflow
90 minAdvanced
In this working session, participants pick a pilot workflow, define the citation-checking standard their team will hold to, and design one governed notebook workflow end to end as the program's capstone.
A pilot workflow with a clear owner
The citation-checking bar your team holds to
A champion who spreads good notebook habits
Time saved, measured after the pilot
A capstone that designs and reviews one workflow

What you will learn

Build a notebook that's bounded
Questions scoped enough to get a grounded answer
Check citations before claims ship
Briefings and Audio Overviews, edited for accuracy
The data and IP boundaries that keep a shared notebook safe
Share a notebook with clear ownership rules
Know when to reach for open-web search instead
Run a pilot with a defined measurement plan

Who should attend

Research and insights teams handling competitive intelligence
Enablement and learning and development teams responsible for internal knowledge
Consultants and analysts working through large document sets
Marketing and strategy teams building briefings from reports
Knowledge-management and documentation leads
Team leads piloting a source-grounded workflow

The program runs one day and splits easily into two half-day sessions. Onsite and live-online delivery use the same syllabus, with labs built around a document set your team researches from.

Format: Onsite or live online
Duration: 1 day (about 8 hours, can be split into two half-day sessions)
Group size: Up to 16 participants per group
Language: English or Turkish
Materials: Citation-checking checklist, notebook-naming template, and a source-boundary worksheet
Certificate: Certificate of completion

Google Workspace & Gemini6 modules · 13 hoursInstructor-led Gemini training built around the tools a team already has open all day: Docs, Sheets, Slides, Gmail, Meet, and Drive. Prompting with useful context, checking sources before anything ships, KVKK-aware data rules, and adoption that survives past the workshop.View training

Perplexity Enterprise6 modules · 13 hoursInstructor-led Perplexity Enterprise training for research and knowledge teams: grounded search setup, a shared Space workflow, sourcing and citation judgment, and enterprise admin and data governance.View training

AI Literacy & Prompt Engineering6 modules · 13 hoursAI literacy and prompt engineering training for your whole company. Teams practice on their own workflows across ChatGPT, Copilot, and Gemini, and walk out with a shared prompt library they use again the next day.View training

AI in Data & BI Analysis6 modules · 12 hoursSQL query drafting, dashboard narratives, data documentation, and stakeholder reporting, taught to data and BI analysts with every number checked against source before it ships.View training

Share which document sets and research questions your team works from. We'll build the labs and session around the citation-checking standard those sources require.

A colleague adjusting a laptop console while comparing a sample with its output

Who ends up in this room?
Research, insights, competitive-intelligence, enablement, and knowledge-management teams that work through large document sets like reports, transcripts, decks, and PDFs every week. Strategy teams fit well too, especially consultants and analysts building briefings from that material. We plan the group and a sample document set together so the labs match current research questions.
How is this different from your Google Workspace with Gemini training?
This program is scoped to bounded, source-grounded notebooks: a defined document set you question, cite, and turn into briefings or Audio Overviews. Google Workspace with Gemini training covers the broader productivity suite instead, Docs, Sheets, Slides, Gmail, Meet, and Drive. The two overlap on prompting and verification habits, but the tools and the workflows are different.
Does this assume any technical skill?
No. The training is for research and knowledge-work staff, and it assumes no coding, data-science, or engineering background. You only need a working knowledge of the documents and questions your team already deals with day to day.
Does this training cover building a custom RAG system?
No. This is a training program on the notebook workflow: sourcing, querying, citation checking, and outputs. If your team needs to build a custom RAG pipeline, that's separate engineering work outside this program's scope.
How do you handle confidential or client data during the training?
Exercises use approved non-sensitive or reduced examples, with synthetic document sets agreed with your team in advance. We don't process confidential or client data in the training environment. Personal data follows the same boundary. The only exception is when your organization has an approved environment and data policy that explicitly allows it.
Can the training be tailored to our research workflows?
It has to be, or the labs won't mean anything to your team. Before the session we run a discovery call to learn your document types and research questions. We also map the stakeholder outputs your team produces. Then we build the labs around a realistic version of that material instead of generic examples.
Does the program promise a certain accuracy or productivity gain?
We don't promise a specific productivity gain or guarantee accuracy or compliance. What we teach instead is the evaluation and citation-checking habit that keeps a human reviewer accountable for every claim a notebook produces. Zeo is not an official NotebookLM vendor partner, and Zeo is not a certified provider of any kind. The habit is the deliverable.
What do teams keep after the training?
A citation-checking checklist, a notebook-naming and organization template, a source-boundary worksheet for data and confidentiality decisions, and a pilot plan with a named owner and a measurement approach for the following quarter.

By team

By industry

By tool

NotebookLM Training for Research & Knowledge Teams