If you’re wrangling data into Proof and something’s not working, you’re not alone. Even the best setups break, especially when you’re juggling APIs, file uploads, and third-party sources. This guide is for anyone who’s hit a wall getting data to sync, show up, or play nice inside Proof. We’ll skip the fluff and get straight to what actually solves these headaches—whether you’re a developer, an ops person, or just the one everyone calls when “the numbers look weird.”
1. Start Simple: Is It Even Connected?
Before you deep-dive into logs or tweak mapping settings, confirm the basics. Integration issues often come down to something dead simple.
Checklist: - Are your source and destination credentials still valid? (APIs expire, passwords change.) - Did someone adjust permissions on your source system? - Is the integration enabled in Proof? - Are you pulling from a sandbox or production environment—and is it the right one?
Pro tip: Don’t trust screenshots from last week. Log in and verify everything yourself.
What to ignore: Fancy dashboards that say “Connected” aren’t always telling the truth. Test with a real data pull or push.
2. Authentication Failures: Credentials, Tokens, and Timeouts
“Authentication error” is the most common (and least helpful) error message you’ll see. Here’s how to break it down:
- Expired API keys or OAuth tokens: Most expire after a set time or if someone resets them.
- Two-factor or SSO changes: If your organization just rolled out new security, old integrations can break.
- IP whitelist mismatches: Your office moved, or you’re connecting from a new cloud provider.
How to fix: 1. Generate new API keys or tokens from the source system. 2. Update the credentials in Proof’s integration setup. 3. If using OAuth, fully disconnect and re-authorize—sometimes “refresh” just doesn’t cut it.
What Proof does well: Proof tends to surface auth failures quickly, usually within the integration settings. Don’t just retry—replace the credentials.
3. Data Mapping Mayhem: Fields Don't Line Up
You set up the integration, but the data looks off. Maybe fields are blank, or numbers don’t match. Nine times out of ten, it’s a mapping problem.
What to check:
- Field names: Are you matching “customer_id” to “clientId” or something else?
- Data types: Is Proof expecting a string and you’re sending an integer?
- Required fields: Did you skip a field that Proof marks as required?
How to fix: - Revisit the data mapping screen in Proof and double-check every field. - Use a sample row—really look at it. Does it land in Proof the way you expect? - If you’re using CSV uploads, open the file in a plain text editor. Excel sometimes reformats data without telling you.
What doesn't work: Guesswork. Don’t just “map everything to everything.” That’s a recipe for bad data and hours of cleanup later.
4. Data Format & File Issues: CSVs, Dates, and Encodings
If you’re importing files, half the headaches come from the file itself.
Common culprits: - Wrong column order: Proof expects columns in a certain sequence. - Date formats: 2024-06-20 vs. 06/20/2024—Proof might only accept one. - Encoding: Non-UTF-8 files can create invisible errors, especially from Excel or older systems.
How to fix: - Open your file in Notepad (Windows) or TextEdit (Mac) to check for weird characters. - Re-save the file as UTF-8, not ANSI or “Mac Roman.” - Check for hidden header rows or blank lines at the top—delete them. - Make sure your date formats match what Proof expects (check docs or the integration setup screen).
Pro tip: If you keep hitting file errors, try uploading a tiny test file with just one or two rows. Fixing problems is way faster on small samples.
5. Third-Party API Headaches: Throttling, Schema Drift, and Timeouts
Pulling from external APIs? Welcome to a world where “it worked yesterday” means nothing today.
Common gotchas: - Rate limits: You’re sending too many requests and getting throttled. - Schema changes: The source added/removed fields and your setup is now broken. - Timeouts: Slow endpoints can cause Proof to drop the connection or skip the sync.
How to fix: - Check the API docs for rate limits. Set Proof to run less often, or add delays between syncs. - If schema changed, revisit your mappings in Proof and update as needed. - For timeouts, see if Proof has “retry” options or suggest pulling smaller data batches.
Honest take: There’s no magic solution here. External APIs change all the time, and you’re always playing catch-up. Build some slack into your schedule for this.
6. Data Not Updating or Syncing Incompletely
The integration runs, but your data’s stale or only half of it shows up. Here’s what to check:
Potential issues: - Incremental sync misconfiguration: Are you only pulling “new” records, but the criteria is off? - Filters applied: Someone (maybe you!) set a filter that’s hiding expected data. - Partial failures: Proof might sync some records, but not all, if there are errors mid-stream.
How to fix: - Review your sync settings—make sure the “last updated” or similar fields are mapped and working. - Remove or adjust filters and re-run the sync. - Check Proof’s sync logs for “skipped” or “failed” rows—often, a single bad record can block the rest.
Pro tip: Always check the logs. If Proof doesn’t provide enough detail, contact their support with the sync ID or error message. Vague “something went wrong” errors are usually a hint to look deeper.
7. Permission & Access Problems
You’re set up, everything looks fine, but nothing moves. It might be permissions.
Check: - Does the Proof integration user have “read” or “write” access to the right tables or endpoints? - Did someone change roles or group assignments in the source system? - Are there IP restrictions or firewall rules blocking Proof’s requests?
How to fix: - Double-check the source system’s audit logs for denied requests. - Ask your IT team if anything changed on the network side. - Make sure the Proof integration user isn’t “read-only” if you’re trying to push data.
What doesn’t work: Just asking “can you check the permissions?” without specifics. Get the exact error message and the user ID Proof uses.
8. When All Else Fails: Debugging Like a Pro
Sometimes you’ve checked everything and nothing makes sense. Here’s how to avoid spinning your wheels:
- Reproduce the problem with the smallest possible test case. Use one record, not a full dataset.
- Check version history: Did someone (including you) change a setting between when it worked and now?
- Document what you’ve tried. If you need to escalate to Proof support, a clear list of steps saves hours.
- Don’t be afraid to start over. Sometimes a fresh integration setup is faster than debugging a broken one.
What to ignore: Online forums full of generic advice. Your setup is unique—copy-paste fixes rarely work.
Keep It Simple: Iterate, Don’t Over-Engineer
Data integrations always sound easy, but even “plug and play” tools like Proof involve trial and error. Don’t stress if something breaks—everyone hits these snags. Focus on getting a minimal, working pipeline first, then improve from there. Iterate, document the weird stuff, and resist the urge to add complexity before you need it. The simpler your setup, the less there is to fix next time things go sideways.