Getting Better AI Analysis Results
How to get the most useful insights from Bit Buddy's AI features.
Bit Buddy's AI is only as good as the data it has to work with. The more complete and accurate your device entries are, the better the analysis and recommendations will be. This guide covers practical ways to improve the quality of the insights you get.
Keep Your Device Data Up to Date
This is the single most impactful thing you can do. If your storage usage numbers are months old, the AI is working with stale data and its suggestions won't be very useful.
Make it a habit to update your used storage amounts periodically. You don't need to do this daily, but a quick check every few weeks keeps things accurate. The Quick Edit feature on a device's detail page makes this fast.
Use Descriptive Names
Device names matter more than you might think, especially when using on-device AI. A device named "Drive 1" gives the AI very little context compared to "Video Projects SSD" or "Time Machine Backup." The AI uses names to understand what role a device plays in your setup, and clearer names lead to more relevant suggestions.
If you're using cloud AI, keep in mind that device names are anonymized before being sent. The cloud provider sees generic labels like "External Drive" instead of your actual names. On-device AI, however, has full access to names, so being descriptive pays off there.
Fill in Content Types
When adding or editing a device, there's a field for content type (what kind of files are stored on the device). Setting this helps the AI understand the purpose of each device. A drive full of video files has different needs than one storing documents or system backups, and the AI adjusts its recommendations accordingly.
Add Tags
Tags give the AI extra context about how devices relate to each other. If you tag a few drives as "Photography" or "Archive," the AI can spot patterns across those groups and make more targeted suggestions. It's a small step that adds up, especially once you have more than a handful of devices logged.
Log Maintenance Activities
The AI pays attention to your maintenance history. If you've been logging backups, cleanings, health checks, and other tasks, the AI can identify devices that are overdue for attention or spot trends in how you maintain your gear.
Even simple entries like "Backed up" or "Ran disk check" are useful. The AI looks at both the timing and the type of maintenance to form its recommendations.
Set Purchase Dates
Adding a purchase date (or at least an approximate one) lets the AI factor in device age. It might recommend replacing aging hardware, adjusting backup frequency for older drives, or flagging devices that are approaching typical end-of-life timelines. Without a date, it can't make those calls.
On-Device vs. Cloud: What Changes?
The quality of analysis can differ between the two providers, and it's worth understanding why.
On-device AI sees everything: names, notes, file contents, locations, and all other details. This means it has the full picture and can give more specific, contextualized advice. The tradeoff is that the on-device model is smaller, so it can sometimes be less detailed in its responses.
Cloud AI works with anonymized data, so it doesn't know your device names, notes, or locations. It works with device types, percentages, tags, and status flags. The cloud model is typically larger and can produce more detailed analysis, but it's working with less context about your specific setup.
Neither option is strictly better than the other. On-device has more data but a smaller model; cloud has a bigger model but less data. In practice, both produce useful results for most users.
Refresh When Needed
AI Insight Cards cache their results for 24 hours. If you've made significant changes to a device (updated storage, added maintenance logs, changed tags), the cached insight won't reflect those updates. Tap the card and use the refresh button to get a new analysis with the latest data.
Quick Checklist
For the best AI results, make sure your devices have:
- Accurate, up-to-date storage usage
- Descriptive names (especially for on-device AI)
- Content type set
- Relevant tags applied
- Recent maintenance logs
- Purchase date (even an approximate one)
You don't need all of these for every device. But the more complete your entries are, the more useful the AI's analysis will be.