James D. Feldman, CSP is a keynote speaker, AI implementation strategist and founder of SOAR Academy. He helps incentive and corporate travel professionals turn the AI tools they already own into competitive advantages. Explore the AI Strategic Concierge at aiconcierge.solutions or reach him at shifthappens.com.

It was a Sunday night. I had three program documents open, a vendor email I hadn’t answered in four days, a half-finished attendee communication that read like a legal disclaimer and a post-program debrief I’d been promising myself I’d write for two weeks.

Sound familiar? Good. Because that Sunday is when I decided to stop carrying all of it in my head and hand it to NotebookLM.

I want to be direct about what NotebookLM is — and what it isn’t. It is not a chatbot. It is not a general-purpose AI assistant. It is a source-grounded research and organization tool. You feed it your documents — RFPs, BEOs, attendee surveys, vendor contracts, program notes — and then you query it like you’d query a brilliant colleague who has read everything and forgets nothing. The key word: it only works from what you give it. No hallucinated data. No generic responses. Your data, your programs, your answers.

I used it for one week across five specific incentive and corporate travel workflows. Here’s what changed.

1. I Stopped Rewriting the Same Vendor Emails From Scratch

Every program has a vendor communication cycle that eats hours: confirmation emails, specification updates, day-of briefings, post-program thank-you notes with outcome summaries. I had been writing these from scratch — or worse, digging through old emails to find a version worth repurposing. I loaded two years of vendor correspondence, program specs and my standard communication templates into a NotebookLM notebook. Then, I queried it.

Copy-Paste Prompt: Based on the uploaded program specifications and past vendor communications, draft a pre-program briefing email to our ground transportation vendor for a three-day incentive trip to Scottsdale for 120 top performers. Include pickup schedule, special accommodation notes for VIP guests, emergency contact protocol and our standard performance expectations. Tone: professional, direct, specific.

Result: A complete, on-brand vendor briefing email in four minutes. Accurate to my actual specs. No starting from a blank page.

This alone recovered about three hours I had been losing every program cycle. Multiply that across your calendar year.

2. I Stopped Second-Guessing My Attendee Communications

Here is the incentive planner’s silent embarrassment: We send communications to top performers — people who earned their spot — and the email reads like a corporate IT notice. Nobody intends it. It just happens when you’re rushed and copying from last year’s template.

I uploaded my current program details, destination information, itinerary and three previous attendee welcome emails into NotebookLM. Then, I asked it to do better.

Copy-Paste Prompt: Using the uploaded program itinerary and destination details, write a pre-trip welcome email to attendees who qualified for the President’s Club incentive trip to the Amalfi Coast. These are top-performing sales professionals who earned this trip. The tone should feel like a personal note from a host who is genuinely excited for them — not a logistics notice. Mention three specific experiences they can look forward to, include essential arrival information and close with one sentence that acknowledges what their achievement means. Under 350 words.

Result: A welcome email that actually felt like a celebration. Specific to the destination, specific to the achievement, specific to what they were about to experience. The kind of email a top performer reads and forwards to their spouse.

3. I Stopped Dreading The Post-Program Debrief

The post-program debrief is the most important document nobody reads. It goes into a folder. The next program begins. The same catering issue happens again. The same AV vendor underdelivers at the same moment in the run-of-show. And everyone acts surprised.

I loaded six years of post-program reports, attendee surveys and client feedback into a single NotebookLM notebook. What came back was not a summary. It was a pattern.

Copy-Paste Prompt: Analyze the uploaded post-program reports and attendee surveys from the last six incentive programs. Identify: (1) the top three recurring operational issues across programs, (2) which destination types and experience formats generated the highest attendee satisfaction, (3) any patterns in client feedback suggesting a systemic process gap, and (4) two specific recommendations for the next program. Present as a one-page executive briefing for a C-suite client.

Result: A one-page executive briefing that identified our top three recurring operational failures, which program formats drove the highest attendee satisfaction scores and two vendor relationships worth re-evaluating. The kind of strategic intelligence that used to require a consultant.

Six years of program data. One query. Thirty seconds. The pattern was obvious once I stopped carrying it all in my head.

4. I Stopped Losing Time To Contract Review

Hotel contracts, DMC agreements, air contracts, ground operator terms — an incentive program can generate 15 to 20 documents with clauses that will cost you money or goodwill if you miss them. Force majeure language. Attrition penalty triggers. Cancellation windows. The terms that matter are buried. I loaded five recent contracts into NotebookLM and asked a single question.

Copy-Paste Prompt: Review the uploaded hotel and vendor contracts for our upcoming incentive program. Flag: (1) all cancellation penalty triggers and their financial exposure, (2) any attrition clauses with thresholds below 85%, (3) force majeure definitions that may be narrower than standard industry language, and (4) any auto-renewal clauses. Summarize findings as a risk briefing I can review with my client before signing.

Result: Every meaningful risk flag in the stack, organized by category, with the specific clause language cited. What used to take two hours of careful reading took eight minutes. And I caught a 90-day notice requirement I had almost missed.

5. I Stopped Starting From Zero On Every New Program Proposal

Every new incentive program proposal involves pulling together destination research, program structure options, experience recommendations and budget frameworks. It’s not creative work. It’s retrieval work. And retrieval work is exactly what AI is built for.

I built a NotebookLM notebook that contains every program I’ve run, every destination fact sheet I’ve ever compiled, every supplier brief and every client objective summary I’ve written. It is, effectively, my institutional memory — searchable in seconds.

Copy-Paste Prompt: Using the uploaded program history, destination profiles and supplier briefs, draft a preliminary incentive program concept for a pharmaceutical client with 85 qualifiers, a $4,500 per-person budget, a preference for culturally immersive destinations and an objective of reinforcing sales team cohesion after a merger. Suggest two destination options with brief rationale, a sample three-day itinerary framework for each and three experience elements that align with the cohesion objective.

Result: A first-draft program proposal structure in 20 minutes. Grounded in real program history. Specific to the client’s stated objectives. Something I can refine and present, not something I’m building from scratch at 11 p.m.

What This Means For Your Practice

I want to be clear about something. NotebookLM did not make me a better incentive planner. The experience, the relationships, the strategic judgment — that’s still mine. What it did was stop draining me on the work that isn’t strategic. The retrieval. The drafting. The pattern-finding in data I already had but couldn’t access efficiently.

The planners I see struggling with AI right now are the ones trying to use general-purpose tools for industry-specific problems. They ask ChatGPT something and get a generic answer that doesn’t know their programs, their clients or their standards. NotebookLM is different because it only answers from what you give it. Your programs become the database. Your standards become the baseline. The AI doesn’t replace your expertise — it runs on it.

The setup takes an afternoon. Load your last three to five programs, your standard communications, your vendor contracts, your post-event reports. Create separate notebooks by client, by program type or by function — whatever matches how you think.

Then start querying.

The week I spent testing this, I recovered roughly nine hours. More importantly, I stopped carrying a dozen open loops in my head that had been quietly draining me for months.

A little AI enhances everything. Too much ruins it. One notebook, one problem, one query. Start there.

NotebookLM is free. The time it gives back is not nothing. |C&IT|