At FitMetrics, our AI Meal Analysis tool is powerful — but the quality of the results depends heavily on how well the photo is taken. Here’s how your clients can get the most accurate food scan possible.
✅ 1. Take a Clear Picture
The #1 rule: make sure the image is clear.
Blurry or dark photos reduce accuracy dramatically. The AI needs to "see" the individual ingredients to analyze them properly. Always recommend your clients take photos with good lighting and clean visibility of all items.
📦 2. Avoid Scanning Packages
Some users take pictures of food still in packaging (protein bars, yogurt cups, etc). While the AI can work with packaging if the client adds a detailed description, it's not as accurate as scanning the actual food.
If your client insists on using packaging, encourage them to add a short description (e.g., “Quest protein bar, 190 kcal, chocolate chip flavor”) to boost accuracy.
🍱 3. Don’t Overlap Food Types
Make sure foods are placed clearly and aren’t overlapping — like hiding rice under meat or stacking items. If the AI can’t see a food, it can’t analyze it. The better the separation between ingredients, the more accurate the result.
💬 4. Forgot to Take a Picture? Use the Chat!
If your client forgets to take a photo, no problem — they can still get their meal analyzed using the built-in AI command.
All they have to do is go to their messages with the coach inside Trainerize and type:
/mealhelp 300g grilled chicken with handful of sweet potatoes and fresh orange juice
The AI will generate the same meal analysis — no picture required.
This feature is perfect for busy clients who forgot to snap a photo but still want accurate tracking.
💡 5. Teach Clients to Use These Features
Most clients don’t know these options exist until you teach them. So make it part of your onboarding process to:
Show how to take the right kind of meal photo
Explain the
/mealhelpchat feature inside TrainerizeShare examples of good vs bad scans