How to Prepare for a Meeting With AI: The Pre-Meeting Brief That Replaces Manual Research

By Sam | SketricGen

Key Points

• Meeting preparation is still mostly manual: scan LinkedIn, skim old emails, guess at the agenda.
• Manual research costs real time. Researching one attendee takes 5-10 minutes. A five-person meeting can eat 30-50 minutes before anyone says a word.
• Subscription "AI meeting" tools mostly help during the call (notetaking, transcripts). Almost none help before it.
• An AI agent can pull your calendar, research attendees and companies, check your Gmail and Slack history, and hand you a finished brief automatically.
• SketricGen's Pre-Meeting Preparation template does exactly this, and you can extend it as your workflow grows.

Who This Is For

• Consultants and client leads prepping for external meetings
• Sales reps and SDRs prepping for calls
• Executives and chiefs of staff prepping for internal or external meetings
• Recruiters and biz dev people prepping for interviews or partnership conversations
• Account and customer success managers prepping for recurring client syncs

Why Meeting Preparation Matters: The Real Cost of Walking In Unprepared

Most people don't skip meeting prep because they don't care. They skip it because the calendar doesn't leave room for it. Back-to-back meetings leave five or ten minutes between calls, not enough time to research anyone properly, so prep either gets rushed or dropped entirely.

Globally, an estimated 24 billion hours a year are lost to unproductive meetings, and roughly a third of the average person's weekly meeting time is considered wasted, according to research compiled by Pumble. In the US alone, Doodle's State of Meetings research put the annual cost of pointless meetings at $399 billion, per a breakdown from Flowtrace. Estimates vary across studies depending on methodology, but the direction holds: meetings that go badly because nobody prepared are an expensive, recurring drain, not an occasional slip.

The sales side has its own version of this math. Chris Beall, CEO of ConnectAndSell, has argued that at typical connect rates, a rep doing five minutes of pre-call research on every prospect burns roughly 100 minutes of research time per successful connect, once every call that never connects is factored in. See his full argument on SalesPOP!. Manual research doesn't scale with a busy calendar. It scales with how much free time you don't have, which is usually very little.

Decision rule: if you have fewer than 3 external meetings a week, a manual checklist is fine. Past that, the time cost compounds faster than most people notice, and automating the research step starts paying for itself within the first week.

Meeting Preparation Checklist vs. AI Notetaker vs. AI-Built Pre-Meeting Brief

What Meeting Preparation Actually Means

Meeting preparation is the work of getting ready for a meeting before it starts: knowing the objective, understanding who's in the room, and having the context you need to contribute instead of spending the first ten minutes figuring out what's going on.

It is not the same thing as an AI meeting assistant like Otter, Fireflies, or Fathom. Those tools listen to the meeting itself and produce transcripts and summaries afterward. They're useful, but they solve a different problem: what happened during the call, not what you needed to know walking in.

Manual Meeting Prep vs. AI Meeting Prep Tools: What People Actually Do

The manual checklist. Research each attendee on LinkedIn, check for recent company news, review the agenda, jot down objectives. This works, but it's slow. BoardEx's client-meeting research guide puts manual attendee research at 5-10 minutes per person, which adds up fast for any meeting with more than one or two people in it.

The subscription point tool. A newer category of tools, like the open-source Brief My Meeting or Cirrus Insight's Meeting AI, will email you a brief before external meetings by reading your calendar and inbox. These are a real improvement over doing it by hand. The limitation: they're single-purpose. They read your calendar and email, and that's it. If your context also lives in Slack, a CRM, or a project tool, you're back to filling the gaps manually.

Neither approach treats meeting prep as a workflow you can shape around how you actually work. Both are fixed: one fixed by your own memory and time, the other fixed by whatever the vendor decided to support.

Meeting Brief Template: What a Good AI-Built Brief Actually Contains

A brief worth reading before a meeting needs more than an attendee's job title. It needs:

• The meeting's objective and agenda context
• Attendee and company research pulled from the open web and recent news
• Internal history: past emails and Slack messages with the same people or company
• A fallback plan for when there's no internal history yet, such as a new prospect or a first meeting

This is precisely how SketricGen's Pre-Meeting Preparation template is built, using the same multi-agent workflow pattern behind most SketricGen templates: one orchestrator, multiple specialized agents. A Meeting Prep Orchestrator agent pulls meeting details from Google Calendar. It hands attendee and company research to an External Research Agent, which runs web search and a news API, and hands internal history duties to an Internal Context Agent that checks Gmail and Slack. If there's no internal history to find, the workflow falls back to public-context research on the meeting topic instead of coming back empty.

How to Build an AI Agent for Meeting Prep, Step by Step

1. Start from the Pre-Meeting Preparation template instead of building the workflow from a blank canvas.
2. Connect Google Calendar so the agent knows what's coming up and who's attending.
3. Connect Gmail and Slack so the Internal Context Agent has something to search.
4. Run it once against an upcoming meeting and review the output before your next one.

Pro tip: the first run against a brand-new contact will look thinner, because there's no internal history yet. That's expected. The fallback kicks in and researches the topic publicly instead. It gets sharper the more history builds up with a given person or company.

You can extend this past the defaults. Notion is supported as an add-on if your team's context lives there too, and SketricGen connects to a wider marketplace of apps if your workflow eventually needs more than calendar, email, and chat. The broader point stands either way: this is a workflow you build and shape, not a subscription you're stuck with.

Common Meeting Preparation Mistakes to Avoid

• Skipping the objective. Attendee research without a clear goal produces a brief nobody reads twice.
• Ignoring internal history. The manual checklist almost never accounts for this, and it's often the most useful part of a brief for a returning client or candidate.
• Assuming a notetaking tool covers this. It doesn't. It solves the during-the-meeting problem, not the before-the-meeting one.

Mistake I see a lot: treating "meeting prep" as one task instead of two. Prep the context before, and let a notetaker handle the record after. Trying to make one tool do both usually means neither gets done well.

Where an AI-Built Brief Falls Short

No brief, automated or not, replaces judgment, and the limits are worth naming plainly.

If all you need is an inbox-only brief and nothing more, a point tool like Brief My Meeting is a legitimate, faster starting option. The agent-based approach pays off once you want prep wired into more of your stack than just email, or once you're running enough external meetings that a single-purpose tool starts feeling limited.

It also shouldn't be visible in the room. On a Reddit thread about what makes a good AI meeting assistant, one commenter argued that the best AI is the one nobody notices: if you have to explain a digital assistant to a client, the tool has already failed. The same logic applies to prep. A brief is something you read before you walk in, not something you mention once you're there.

And it works best once you already know what good prep looks like. On a Reddit thread about pre-call research, one rep argued for doing the research manually first: build the workflow by hand, then automate it, because manual research surfaces nuance a template can miss. That's a fair caution. An AI-built brief works best as a replacement for repetitive research you already understand, not a substitute for learning what to look for in the first place.

What Practitioners Are Saying on Reddit and Quora

Across Quora threads asking how people prepare for important meetings, the same manual pattern shows up repeatedly: a quick LinkedIn scan of attendees, a review of the agenda, and a mental run-through of the goal 10-15 minutes beforehand, per the discussion on Quora. It's a reasonable process, and it's entirely manual.

Reddit tells the same story with more detail. One rep new to sales, coming from a machine learning background, asked how everyone else was finding time to research this much before every call: checking LinkedIn, podcasts, old blog posts, and company news before every single conversation. They ended up building themselves a research agent to keep up, adding that they couldn't imagine doing it by hand, opening five tabs for every single prospect.

Why it works: One reply on the same thread put it simply: doing the homework changes how people see you once they realize you know exactly who you're talking to. See the thread.

A separate thread on wasted SDR research time surfaced a more specific version of the same problem. One commenter described spending 10-15 minutes per account digging through the CRM, call recordings, LinkedIn, and news searches, then switched to pulling all of that into one view before a call: prep went from 15 minutes to about 3, because the real problem was never the research itself. It was that the information was fragmented, and reps were the ones manually connecting the dots every time. That's the exact gap an Internal Context Agent checking Gmail and Slack is built to close.

Author Take - Sam

Most teams don't need a better meeting-prep checklist. They need to stop doing the checklist by hand. My rule of thumb: if you're repeating the same research steps for every external meeting, that's not diligence, it's a workflow you haven't automated yet. The fix isn't a longer checklist. It's letting an agent run the checklist for you before you even open your calendar.

Next Steps

Manual checklists and inbox-only tools both leave gaps. Try the Pre-Meeting Preparation template to see the full brief an agent can put together before your next meeting, or explore SketricGen's app dashboard to build a version tuned to your own stack.

Related reads

• How to Build an AI Sales Assistant for Lead Qualification
• How to Build a Multi-Agent Workflow in Under 10 Minutes
• AI Agent for Lead Generation: The 2026 Playbook