More than half of consumers say they still have the last promotional product they received, and nearly two-thirds say they clearly remember the brand that gave it to them. That should tell every event team something important right away: merch is not the small extra. It is often the thing that lingers after the booth comes down, the badges come off, and the follow-up emails get buried. When you plan it with real data instead of gut instinct, branded gifts stop feeling random and start feeling personal, useful, and honestly hard to throw away.
What “AI-Powered Merch Planning” Really Means for Event Teams
Let’s make this simple. AI-powered merch planning does not mean letting a robot pick your swag and hoping for the best.
It means using the data you already have, plus a little smart analysis, to make sharper product decisions before you place the order. Think attendee role, age range, climate, event format, travel behavior, ticket tier, sponsor level, past event feedback, and even what people actually carried, wore, or posted last time. Then AI helps you spot patterns faster than a human spreadsheet marathon ever could.
For event teams, that can look like:
- grouping attendees into useful personas
- predicting which product categories fit each segment
- forecasting how many of each item you really need
- reducing sizing mistakes for apparel
- building smarter VIP, speaker, sponsor, or staff kits
- avoiding the classic overbuy on items no one wanted in the first place
The real win is not “more tech.” It is better judgment at scale.
A good AI setup should still be guided by human taste, brand fit, and common sense. That matters because swag lives in the real world. It sits on desks, gets packed into carry-ons, shows up in dorm rooms, rides in gym bags, and ends up in everyday routines. That is why event merch works best when it feels less like a giveaway and more like something someone would have picked for themselves.
This is also where great sourcing partners matter. JNP lives in that sweet spot where trend awareness meets real-world execution. If you already know your audience loves tactile, high-visibility products, something like custom football towels can hit hard for athletics, campus events, and spirit-driven activations without feeling overworked.
The #1 Metric That Predicts “Keep Rate” (And How to Track It)
If your team only tracks cost per item, you are missing the metric that actually matters most: keep rate.
Keep rate is the percentage of recipients who still use, wear, carry, or keep the item after the event window closes. It is the closest thing to a truth test for whether your merch was a hit or just table décor.
Here is a clean way to define it:
Keep Rate = Number of recipients still using or keeping the item after a set period / Number of recipients who received it
You can measure it 30, 60, or 90 days after the event. For bigger annual events, 60 days is usually a sweet spot. It gives people enough time to naturally keep using the product, or quietly stop.
Track it with a mix of methods:
- post-event survey questions
- social listening for organic usage photos
- repeat-wear sightings for apparel
- QR-based product feedback
- team observations from VIP and sponsor follow-up
- reorder patterns for internal events or recurring conferences
A smart pro tip: do not ask only “Did you like the merch?” Ask better questions.
Try:
- Which item have you used more than once?
- Which item would you choose again?
- Which item did you keep in your bag, office, car, or home?
- Which item felt most worth the brand giving you?
That is where the signal lives.
If you want a better feedback loop, look at a guide to post-event surveys from Qualtrics and build merch-specific questions into your standard post-event wrap-up. The merch teams who improve fastest are not always the ones with the biggest budgets. They are the ones actually measuring what got used.
What Data to Pull Before You Pick Products
Before you choose product categories, pull the data first. Not after someone on the team says, “Everybody loves hoodies.”
Start with what is already available inside your event ecosystem:
- registration data
- attendee job titles or roles
- geography and climate
- whether guests are flying or driving
- ticket tier or access level
- sponsor versus attendee versus speaker segmentation
- past event survey feedback
- booth traffic or session selection data
- event type, whether campus, corporate, gala, fundraiser, or sports-centered
This is where your registration flow becomes surprisingly powerful. If your platform allows custom registration questions, use them. Tools like Zoom’s registration customization features show how event teams can collect custom fields, require certain answers, and track registration source data. That is not just for attendance ops. That is merch intelligence.
For example, if your registrants are:
- student attendees, you may lean wearable, social, and budget-smart
- executives, you may lean elevated, useful, and discreetly branded
- developers, you may prioritize desk utility, comfort, and portability
- gala guests, you may want premium gifting with presentation value
The key is not collecting more data just because you can. It is collecting the right inputs so your product choice feels obvious.
Using AI to Match Product Types to Audience Personas
This is where planning gets fun.
AI can help identify patterns between audience traits and product categories, but the merch still needs taste. A 24-year-old attendee at a campus leadership conference and a 54-year-old healthcare executive may both appreciate quality, but not in the same format.
Think in personas, not one giant audience blob.
Campus and Gen Z-leaning events
Apparel, consumables, fashion accessories, and products with social-photo energy tend to land well. These audiences usually respond to items they can wear, show, or immediately work into everyday life.
Executive or sponsor audiences
Premium desk items, upgraded drinkware, refined kits, and lounge-worthy gifts usually outperform loud novelty. This audience wants polish and utility.
Sports, alumni, and spirit-driven events
High-emotion products matter here. Towels, totes, tees, and celebration pieces can feel less like swag and more like identity markers. A project like JNP’s Wesleyan University championship totes and shirts case study is a great example of merch built around pride, memory, and use, not filler.
VIP lounges and hospitality spaces
Comfort-forward gifts can absolutely work when the setting supports them. Think tactile, cozy, and elevated. That is why products like pillows for events can make sense in the right environment, especially where experience design is part of the brand story.
One helpful outside reality check: McKinsey’s research on personalization expectations shows people increasingly expect relevance, not generic treatment. Merch follows that same rule. Relevance wins.
Budget Optimization: Spend More Where It Gets Used
Not every recipient needs the same spend level. Actually, one of the smartest moves in event merch is accepting that uniform budgets often produce uniformly forgettable results.
A better model is tiered allocation.
Here is a simple framework:
Base tier
For all attendees. Keep it useful, easy to pack, and brand-right.
Mid tier
For speakers, sponsors, returning clients, or student leaders. Add better material quality or a second item that deepens perceived value.
Premium tier
For VIPs, partners, media, top donors, or high-stakes prospects. This is where curation matters most.
The point is not to spend more everywhere. It is to spend more where repeat use, visibility, and relationship value are highest.
Also, stop looking only at unit cost. Evaluate:
- likelihood of repeated use
- fit with attendee context
- ease of transport
- perceived quality
- brand alignment
- whether the item earns a second life after the event
Cheap merch is expensive when it gets tossed by sunset.
Size and Fit Predictions for Apparel (So You Don’t Waste Inventory)
Apparel can be a star, or a warehouse mistake.
If you are ordering tees, hoodies, quarter-zips, or outerwear, do not just default to a generic size curve. Use actual historical data where possible. If your event has happened before, compare size requests, sell-through, and leftovers by audience type. If it is a new event, use the closest analog you have.
Helpful inputs include:
- registration gender data, if appropriate and voluntarily provided
- region and climate
- age range
- previous apparel redemption data
- whether the item is fitted, relaxed, or oversized by design
- whether attendees are likely to layer it
A few real-world rules still matter:
- unisex does not mean universally flattering
- premium blanks and soft fabrics usually improve keep rate
- if attendees must fly home, bulkier items need stronger value to justify packing
- redemption-based apparel programs can reduce overordering
When possible, collect size preference during registration. It is better than guessing, and much better than ordering 200 mediums because the team got nervous.
Personalization at Scale: Names, Roles, Tracks, and VIP Tiers
Personalization is where merch starts to feel like a gift.
And no, this does not mean putting everyone’s first name on everything. That can get weird fast.
Smart personalization can include:
- role-based kits
- track-specific inserts
- campus house or team identifiers
- speaker-only upgrades
- sponsor lounge exclusives
- discreet VIP packaging
- initials or names only where the product genuinely benefits from it
The best personalized merch feels considered, not over-engineered.
Ask yourself:
- Does personalization increase usefulness?
- Does it increase emotional value?
- Does it make the recipient more likely to keep it?
- Does it still look good if photographed?
If the answer is no, skip it.
Avoiding “Data Traps”: Bias, Overfitting, and Trend Chasing
This is the part too many teams skip. Just because AI gives you a recommendation does not mean the recommendation is smart.
Bad inputs create bad merch plans.
Watch for these traps:
- Bias: If your historical data only reflects one segment well, your recommendations may ignore everyone else.
- Overfitting: If one product crushed at one event, that does not mean it belongs at every event.
- Trend chasing: Viral does not always mean durable. Some items look hot for two weeks and dead by ship date.
- Over-personalization: Too much data use can feel intrusive instead of thoughtful.
- Brand mismatch: Just because attendees might use an item does not mean it represents your brand well.
A good check here is to stay grounded in trusted guidance on responsible AI and privacy, like the NIST AI Risk Management Framework and the FTC’s guidance on minimizing data. In plain English, use AI to support judgment, not replace it. Keep data collection purposeful. Be transparent internally. And do not turn your merch strategy into a creepy surveillance project.
A Simple AI-Ready Workflow for Event Planners
If you want a practical system, here is one that works:
- Set the merch goal first.
Awareness, retention, sponsor visibility, VIP delight, team spirit, or post-event usage. - Pull the right data.
Registration info, prior survey feedback, audience segment, ticket tier, geography, and historical merch performance. - Build 3 to 5 audience personas.
Keep them behavior-based, not overly theoretical. - Use AI to shortlist product categories.
Not final items. Categories first. - Filter through brand taste and logistics.
Shipping, storage, minimums, packability, decoration method, and timeline still matter. - Assign spend by tier.
Do not flatten your budget. - Test personalization carefully.
Add it where it increases value, not just novelty. - Track keep rate after the event.
Then feed that data into the next planning cycle.
That last step is where most teams level up. Every event should make the next one smarter.
Data Makes Merch Feel Like a Gift, Not a Giveaway
The best event merch is not random, loud, or chosen because someone liked it at another conference six months ago. It is chosen because it fits the audience, the moment, and the way people actually live. AI makes that process faster and more precise, but taste, restraint, and experience still do the heavy lifting.
At JNP, that is the whole point. We are not here to flood a table with throwaways. We are here to help brands choose pieces people want to keep, wear, carry, and remember. When the data is good and the product fit is right, merch stops acting like an expense and starts acting like momentum.
Takeaway: Start measuring keep rate, segment your audience before you source, and let AI help you narrow the field, not make the final call. That is how branded gifts start feeling personal, premium, and worth holding onto.
