MakerWorld, Bambu Lab’s 3D model platform, runs a new design contest almost every week: Halloween decorations, garden tools, RC battleships, board games. Since October 2023 these contests have received 131,899 models and 43.7 million views. As a business analyst, I wanted to find out which themes activate the community, which ones don’t, and what a contest planner should do with that information.
👉 Explore the interactive report
👉 See the Python notebook
The approach
- Collect: gathered all 158 contests from the public contests page, plus each contest’s detail page (prizes and winners).
- Classify: put every contest into one of six theme categories: Seasonal & Holidays, Home & Functional, Toys & Games, Art/Décor & Fashion, Engineering & Tech, and Education & Social Good.
- Score: built a Composite Performance Score (40% participants, 30% models per day, 30% views, each converted to a percentile) and ranked the 150 finished contests into S/A/B/C tiers.
- Analyze: compared the categories, host types (official vs. creator-hosted) and seasonality, and looked at how prize size relates to participation.
What I found
- 🎃 Holidays rule. Twelve Halloween, Christmas, Valentine’s and Easter contests brought in 33.7% of all models submitted, with 4.4x the median participation of other contests. The best contest overall was Cozy Christmas 2024 (6,254 models, 3,362 participants).
- 🗓️ Holiday cannibalization. Regular contests launched from September to December get 34% fewer participants (median 347 vs. 526).
- ⚙️ Engineering gets viewers but few entries. Engineering & Tech contests get about as many views as other categories but have the lowest participation (288 median participants). Only 7% of them reach Tier S/A. The worst contest overall, Automobile, got just 42 models.
- 🏠 Home & Functional is the reliable workhorse. It’s the biggest category (40 contests), and everyday-problem themes like Gardening Tool consistently draw 500+ makers.
- 🧸 Toys & Games produce the most-loved winners, with about 2,300 likes per winning model, the highest of any category.
- 🤝 Creator-hosted and brand-sponsored contests lag, mostly because of visibility (about 159k median views vs. 283k for official contests). Prize size doesn’t predict participation (ρ = −0.18).
Recommendations
- Build the calendar around the four holiday flagship contests.
- Protect regular contests from September to December: run fewer of them, use easier briefs, or extend their deadlines.
- Fill the rest of the year with Home & Functional problems.
- Lower the barrier for engineering contests with starter templates and beginner prize tracks.
- Give creator-hosted and brand contests more promotion instead of bigger prizes.
Python (pandas, matplotlib) · JavaScript for data collection · Chart.js dashboard · KPI design and weighted scoring · segmentation · seasonality analysis · business recommendations
This is the second project in my “game and community analytics” series. The first was a Pokémon Top Trumps competitive analysis.
Data collected from public MakerWorld pages on September 25, 2026. This is an independent study, not affiliated with Bambu Lab or MakerWorld.