Loading tutorials…
Loading tutorials…
Your flow worked yesterday. Today it returns 429 'Too Many Requests.' Power Automate connector limits are real and per-license-tier. This walks the diagnostic sequence specialists run.
Who this is forAnyone running Power Automate flows that started failing with throttle errors, rate-limit messages, or license-tier warnings. Common in flows that scaled past their initial design point.
What you'll need
Step 1
Connector limit vs license-tier limit vs Dataverse API limit. Error messages differ. Diagnose first.
Open the failed run. Read the error message carefully.
Common error patterns:
- "Throttled" or "429 Too Many Requests" → connector-level rate limit hit.
- "API call limit exceeded for this license" → Power Automate Per User/Flow daily limit.
- "Dataverse API limit reached" → Dataverse per-user-per-minute (~6,000) limit.
- "License does not allow Premium connector" → wrong license tier for the connector being used.
Each limit has different fixes — diagnose first, fix second.
Step 2
Power Automate Per User: 6,000 API requests/day per user. Per Flow: 15,000/day per flow. Process license: higher.
Power Automate Per User license: 6,000 daily API requests per user. Sufficient for personal automation.
Power Automate Per Flow license: 15,000 daily API requests per flow. Sufficient for most departmental flows.
Process license: 100,000+ daily API requests. Required for high-volume enterprise flows.
Microsoft Dataverse separate: ~6,000 requests per user per minute (sometimes more for service principals).
Check Power Platform admin center → Analytics → API requests for trends.
If you are consistently near the cap, upgrade the license OR redesign for fewer API calls (next step).
Step 3
Most flows make 3-5x more API calls than needed. Audit each action — can it be batched, cached, or removed?
Audit the failing flow. Count API-consuming actions per execution.
Common optimizations:
- Replace multiple "Get a row" calls in a loop with one "List rows" with $filter — one API call instead of N.
- Use ChangeSet for bulk Dataverse updates — 1,000 updates in one API call instead of 1,000.
- Cache reference data in a Compose variable at the start of the flow — reuse instead of re-fetching.
- Remove redundant lookups (e.g., fetching the same lookup twice in different branches).
A well-optimized flow typically uses 30-50% fewer API calls than its first draft.
Step 4
Wrap throttle-prone actions in retry logic with exponential backoff. Power Automate has built-in retry policies.
For known-throttle-prone connectors (Outlook, Teams, SharePoint), enable retry policy: action Settings → Retry Policy → Exponential.
Default: 4 retries. For high-throttle scenarios, increase to 8-10 retries.
For custom HTTP actions: check the response status. If 429, parse the Retry-After header and wait that long before retrying.
Add Concurrency Control on Apply to Each loops: action Settings → Concurrency Control → set degree to 1-3 to throttle yourself before the connector does.
Step 5
SharePoint REST API has higher limits than SharePoint connector. Custom HTTP can bypass connector throttles (with care).
Standard SharePoint connector has connection-level throttles. For high-volume SharePoint reads, use SharePoint REST API directly via HTTP action (Premium).
For Microsoft Graph (Mail, Calendar, OneDrive): the Microsoft Graph connector has different limits than individual app connectors.
For Dataverse: when individual actions are throttled, ChangeSet operations have higher effective throughput.
For external APIs: custom connectors give you control over retry/throttle behavior that built-in connectors don't.
Step 6
If a single user/flow keeps hitting limits, split the work. Multiple service principals can multiply your effective API quota.
Per User license cap: 6,000 API requests/day per user. If one user is the bottleneck, create multiple service principals.
Set up flows to use different connections (different application users). Each app user has its own API quota.
For round-robin distribution: use a "Get Random Number" action to pick which connection to use per execution.
This pattern is more complex but can effectively 5-10x your API throughput without buying Process license.
Step 7
Power Platform admin center → Analytics → API requests. Watch for trending-up patterns to predict when you will hit caps.
Power Platform admin center → Environments → Analytics → API requests.
Charts: daily API requests per environment, top consumers (users + flows).
If consumption is trending up 20%+ month-over-month, you will hit license caps within 2-3 months at current pace.
Use trend data to plan: license upgrade, flow optimization, or work distribution.
Common mistakes
Not knowing the license-tier API caps
What goes wrong: Build flows assuming unlimited API calls. Hit the 6,000/day Per User cap. Flows throttle for the rest of the day. Critical work backs up.
How to avoid: Know your license tier limits. Per User = 6K/day. Per Flow = 15K/day. Process = 100K+/day. Match flow design to license tier.
API calls inside Apply to Each loops without concurrency control
What goes wrong: Loop over 5,000 records. Apply to Each fires API calls in parallel by default (degree of parallelism = 20). 5,000 records × 3 API calls each = 15,000 calls in 5 minutes. Connector throttles for hours.
How to avoid: Set Concurrency Control on Apply to Each. Degree of 1-3 for throttle-sensitive operations. Or use ChangeSet for bulk operations.
No retry policy on throttle-prone connectors
What goes wrong: Flow hits a transient 429. Default behavior: action fails. Flow halts. Downstream work doesn't happen.
How to avoid: Enable retry policy with exponential backoff. 4-10 retries typically handle 90% of transient throttling.
Re-fetching the same data multiple times
What goes wrong: Flow fetches the same lookup record 5 times in different branches. 5 API calls when 1 would do. Multiply by 1,000 executions = 4,000 wasted API calls/day.
How to avoid: Cache reference data in Compose variables at the start. Reference variables instead of re-fetching.
Ignoring API usage trends
What goes wrong: Monthly API consumption trends up 30% MoM. You do not look at trends. Hit the cap one Tuesday morning. Production flows halt. 4 hours of downtime + scramble.
How to avoid: Monthly review of Power Platform Analytics → API requests trend. Plan license upgrade or optimization before hitting caps.
Recap
Done — what's next
How to set up Power Automate cloud flows
Read the next tutorial
Hand it off
Connector limit issues compound: a small flow inefficiency becomes throttling at scale. EverestX Power Automate specialists optimize for API economy by default — typical first audit reduces API consumption 50-70% without affecting business outcomes. $400-1,000/mo for ongoing.
See specialist rates
Connector throttling: per-connector per-minute limits set by Microsoft (varies by connector, e.g., SharePoint: ~600/min). License-tier limits: daily quota set by your Power Automate license (Per User 6K, Per Flow 15K). Different problems, different fixes.
Power Platform admin center → Environments → Analytics → API requests. Shows daily consumption by user, flow, and connector. Identify top consumers to focus optimization.
No. License-tier daily limits go up (Per User → Per Flow → Process), but connector-level per-minute throttles still apply. Optimization (fewer + smarter API calls) is necessary regardless of license.
Yes. Application users (service principals) have their own API quota. Multiple service principals = additive quota. Common pattern for high-volume flows that exceed per-user limits.
The flow run history shows which action failed with the throttle error. Use the action name to identify the connector. Cross-reference with Microsoft documentation for per-connector limits.
Microsoft Power Automate
Power Automate is the right call when your stack lives in Microsoft 365 — but the licensing maze and Premium connector trap catch most newcomers. This walks the right setup path end-to-end, in plain language.
Microsoft Power Automate
Power Automate disables flows after 14 days of consecutive failures (default). By the time you notice, weeks of automation are gone. This walks the error-handling pattern that catches problems early.
Microsoft Power Automate
Dynamics 365 is the gold-standard automation target — deep schema, rich relationships, complex security. Power Automate is the right tool when it works. This walks the patterns that make it work.
Microsoft Power Automate
DIY Power Automate is great until you have 15 flows, Premium licensing confusion, and a SharePoint admin asking why their list keeps getting cluttered. This is the honest framework: when the cost of self-managing exceeds the cost of hiring.
Zapier
Your Zap was working last week. Today, Zap History shows red. This walks through the diagnostic flow specialists run — OAuth, payload shape, rate limits, schema drift — in the order that surfaces the issue fastest.