Troubleshooting and Important Considerations for PMA MCP Server
Last updated
Note: This article covers known limitations and current behaviors of the Power My Analytics MCP server, followed by specific troubleshooting steps for the issues most commonly encountered. The Limitations and Important Considerations section explains the "why" behind each issue so you can set accurate expectations and avoid unnecessary troubleshooting. The Troubleshooting section follows with Symptom, Cause, and Solution for each specific issue, with a reference back to the relevant limitation where one applies. For setup instructions, see Connect to the PMA MCP Server from Claude Desktop or Connect to the PMA MCP Server from ChatGPT. For a full tools inventory, see the PMA MCP Tools Reference.
Note: The PMA MCP server is now in Beta and available to all PMA users. No early access request is required.
Important: MCP usage is included with your PMA plan and draws from the same shared per-plan limit pool as the Power My Analytics API (each plan includes an API rate limit, a request quota, and a row quota per billing cycle; on the Custom plan the request and row quotas are negotiated with your sales representative). See Get Started with Power My Analytics API for the current per-plan limits.
Limitations and Important Considerations
The behaviors described in this section are known characteristics of the current release, not unexpected bugs. Many will be resolved or superseded as the MCP server moves toward general availability. Where a limitation has a corresponding troubleshooting fix, the Troubleshooting section below provides the resolution steps.
Access and Plan Eligibility
MCP is available to all PMA users during Beta. No early access request or allow-listing is required. Tokens are created automatically when you connect an AI client via the OAuth flow.
MCP access is included with every PMA plan. Your MCP usage draws from the same per-plan limits as the Power My Analytics API, described in the note at the top of this article.
Authentication and Token Behavior
AI clients connect through sign-in (OAuth). Claude Desktop, Claude Web, Claude Code, ChatGPT, and other MCP clients authenticate via OAuth 2.1 + PKCE against your existing PMA login, and the sign-in creates a token for that client automatically. API keys from MCP & API > REST API can't be used to connect to the MCP server; since August 15, 2026, the server rejects them with a 403 error. For a backend service with no browser (such as a scheduled job or a server-to-server integration), use the Power My Analytics API with an API key instead. Either https://mcp.powermyanalytics.com or https://mcp.powermyanalytics.com/mcp works for any client; the two are interchangeable aliases for the same server, so the address you use is never the reason a connection fails.
Your access token and your AI client's registration have different lifetimes. OAuth-issued access tokens are issued at connection time and expire 1 year later. Separately, your AI client's registration with the PMA server lasts 30 days and renews automatically each time you use the connection, so a connection in regular use needs no attention. PMA does not issue refresh tokens, so a registration that ages out (because the connection went unused for more than 30 days, or more than 30 days passed between adding the connector and finishing the sign-in) cannot renew itself: remove the connector and add it again to register fresh (see the entry below, "Authorization Successful, but your AI client then says it cannot connect"). Clicking Reconnect reuses the old registration, so it won't fix this. When an access token expires (or if your hub access is revoked), disconnect and reconnect to mint a new one.
One hub per connection. Customers with access to multiple hubs cannot toggle between them within a single session; to switch hubs, disconnect the connector and reconnect, choosing the desired hub during the OAuth flow.
Data and Query Behavior
The historical-data window cap has not yet been decided. The MCP server does not currently enforce a maximum lookback period. Responses for very old date ranges should be treated as best-effort until the cap is formally set and documented.
Response shapes for high-frequency queries are still being standardized. During Beta, the same prompt may occasionally return data in slightly different structures across users or sessions. Canonical response formats for common queries (sources, accounts, summaries) are being finalized.
The MCP server queries PMA's warehoused data layer, not live platform APIs. Data reflects the state of PMA's warehouse after the most recent successful Data Sync for each platform. Each data source syncs once a day at its sync time (3 AM in the source's time zone by default), on every plan; the Hourly Refreshes add-on refreshes Google Sheets reports and data exports, not the warehouse. Each account's last_sync timestamp is queryable via pma_list_data_sources for current freshness.
Zero-value days are reported as gaps, not anomalies. pma_detect_anomalies explicitly excludes zero-value days from anomaly detection and reports them separately in the data_gaps field. If a customer asks "why didn't it flag the day I had zero spend," this is intentional behavior, not a bug.
ChatGPT-Specific Behavior
ChatGPT connects through developer mode. To use PMA in ChatGPT, you turn on developer mode and create a custom app in ChatGPT on the web; in ChatGPT Business, Enterprise, and Edu workspaces, an admin must allow developer mode first. A newly created app is available only in chats started after you create it. To make sure ChatGPT uses your PMA data, type @ in your message and select the app (for example, Power My Analytics). See Connect to the PMA MCP Server from ChatGPT.
Finish the ChatGPT sign-in soon after creating the app. If more than 30 days pass between creating the app and completing the PMA sign-in, the app's registration with the PMA server expires and ChatGPT shows "Unknown Application." A completed connection isn't affected; it keeps working for the life of its 1-year access token.
Cost and Performance Considerations
Vague prompts cause the AI to make multiple discovery calls before answering. When a prompt does not specify the platform, metric, date range, or analytic intent, the AI must call pma_list_data_sources and related tools before it can begin analysis. Each additional tool call adds latency and, on most AI clients, token cost. Specifying these four elements up front materially reduces the number of tool calls required. See Tips for Writing Effective Prompts in the PMA MCP Tools Reference.
Important: Rate limit: a per-plan rate limit shared between MCP and the PMA API (Free 15, Single (legacy) 15, Business 30, Grow 45, Pro 45, Custom 60 requests per minute). This is a hard limit. A shared per-plan quota of requests per billing cycle also applies (Free 100; Single (legacy) 500; Business 5,000; Grow 15,000; Pro 25,000; Custom negotiated with your sales representative). Agentic AI clients in chatty loops (or analytical investigations that fan out across many platforms in quick succession) can hit this limit and receive a generic "too many requests" error. This is intentional, not a bug. If the limit is hit, pause for a minute before retrying, and consider tightening prompts to reduce the number of tool calls.
pma_inspect_org_data is slow. A full diagnostic scan can take 30–45 seconds. Reserve it for troubleshooting; pass a connector_type to keep individual scans fast.
Unavailable Data
The following are not currently accessible through the PMA MCP server:
- Real-time, non-warehoused data. The MCP server queries PMA's warehoused data layer only, which updates once a day at each data source's sync time.
- Multi-hub session. Within a single OAuth session, the customer can only query one hub at a time. Switching hubs requires reconnecting.
What does this mean for me?
- MCP is included with your PMA plan, and its usage shares your plan's limits with the PMA API.
- Connect AI assistants through sign-in. Scripts, scheduled jobs, and other integrations without a browser use the PMA API instead.
- A connection you use regularly keeps working. If you haven't used it for more than 30 days, remove the connector and add it again.
- Answers come from your most recent Data Sync, and specific questions get faster answers while using fewer requests.
Troubleshooting
Each issue below includes the symptom as the customer would experience it, the cause (with a reference to the relevant limitation above), and the resolution steps.
Connection and Authentication Issues
Authorization Successful, but your AI client then says it cannot connect
Symptom: You finish signing in and see "Authorization Successful!", but your AI client's Connectors panel then shows the PMA entry against https://mcp.powermyanalytics.com with "Couldn't connect to the server. Check that the URL points to a valid MCP server" and a Reconnect button.
Cause: The sign-in (OAuth) finished and a token was issued, but the connection step right after it did not complete. The usual reasons, in the order worth checking:
- The authorization timed out. The pending sign-in is valid for about 10 minutes; a long pause between authorizing and the client connecting lets it expire.
- The client's registration went stale. Your AI client's registration with the PMA server lasts 30 days and renews each time you use the connection. If the connection goes unused for longer than that, the registration ages out, and because PMA does not issue refresh tokens, it cannot recover on its own. Clicking Reconnect will not clear it; the connector has to be removed and re-added.
- A corporate network or proxy is in the way. The sign-in page is ordinary web traffic and can go through, while the streaming connection your AI client uses right afterward is blocked or cut short.
- Your plan's per-minute request limit was hit (a 429 response), your subscription is not active (a 402 response), or a manually created API token was used instead of the one from the sign-in (a 403 response).
Solution:
- Remove the PMA connector from your AI client, add it again at https://mcp.powermyanalytics.com, and complete the sign-in through to "Authorization Successful!" without a long pause. This registers your client fresh and issues a new token, which clears the timeout and stale-registration cases (by far the most common).
- If it still fails, try a different network (for example, off the corporate VPN or on a personal hotspot). If it connects there, the corporate proxy or firewall is the cause: ask your IT team to allow https://mcp.powermyanalytics.com and to let long-lived streaming responses through.
- Confirm the connection registered on the PMA side: open the MCP page in your Hub (left navigation) and check the Connected Application table. If your AI client is listed, the sign-in worked and the fix is on the connection or network side (steps 1 and 2). If nothing appears after several tries, contact support using the details under When to Contact Support.
Note: Remove-and-re-add will not help in three cases. If your subscription is inactive, restore billing first; if you were rate limited, wait about a minute and retry; if a manually created API token was pasted in, remove it and connect through the normal sign-in instead. Both https://mcp.powermyanalytics.com and https://mcp.powermyanalytics.com/mcp reach the same server and either works, so the address itself is not the problem.
Note (existing connections): If you first connected before PMA introduced per-connection tokens and per-user usage visibility, sign out of the connector and reconnect once so those changes take effect.
Your subscription is inactive (402)
Symptom: The connection or your tool calls fail, and your PMA subscription is past due, canceled, or otherwise not active.
Cause: The MCP server returns a 402 response when billing is not active for your PMA account. This is the shared MCP and API access being gated, not a connector bug.
Solution: Restore your subscription in your Hub's billing settings, then reconnect the PMA connector. Removing and re-adding the connector will not restore access while billing is inactive.
Token expired or sudden loss of tool access (browser-based AI clients)
Symptom: The PMA MCP was working, but Claude or ChatGPT can no longer return data from your Hub. Tool calls fail silently or the AI states it does not have access.
Cause: OAuth-issued access tokens are valid for 1 year, but they can expire earlier if your hub access is revoked or your PMA login changes.
Solution: Disconnect the PMA MCP connector in your AI client and reconnect. The OAuth flow re-runs and issues a new token. Refresh tokens are not currently issued.
Connection fails entirely with no clear error
Symptom: The connection attempt fails without a detailed error message. The PMA MCP connector does not appear as connected after completing the setup steps.
Cause: The server address was mistyped. A trailing slash or the /mcp suffix is not a problem; https://mcp.powermyanalytics.com and https://mcp.powermyanalytics.com/mcp are interchangeable.
Solution: Verify that the server URL entered is exactly https://mcp.powermyanalytics.com (a trailing slash or the /mcp suffix works the same) and retry. If the address is correct and the connection still fails right after "Authorization Successful!", see the entry above. If the issue persists, submit a ticket.
ChatGPT-Specific Issues
ChatGPT answers without using any PMA tools
Symptom: You ask ChatGPT a question about your marketing data, but the response is generic, doesn't reflect your Hub data, or says no data is available, even though you've created the Power My Analytics app.
Cause: The chat was already open when you created the app, or your message didn't @mention it (see ChatGPT-Specific Behavior above).
Solution: Start a new chat, type @, select Power My Analytics (or the name you gave the app), and re-send your prompt. If a follow-up answer doesn't reflect your Hub data, @mention the app again in that message.
The Developer mode toggle is missing, or ChatGPT says "Your admin has not enabled developer mode"
Symptom: You can't turn on developer mode in ChatGPT Settings > Security and login, so you can't create the Power My Analytics app.
Cause: ChatGPT Free doesn't include developer mode, or your ChatGPT Business, Enterprise, or Edu workspace admin hasn't allowed it.
Solution: Use a supported ChatGPT plan (Plus, Pro, Business, Enterprise, or Edu). In a workspace, ask your ChatGPT admin to allow developer mode in Workspace settings > Permissions & roles > Connected Data.
ChatGPT shows "Unknown Application" when you sign in
Symptom: After you select Create, the PMA sign-in shows "Unknown Application" instead of completing.
Cause: More than 30 days passed between creating the app and finishing the PMA sign-in, so the app's registration with the PMA server expired.
Solution: Remove the app from ChatGPT (Plugins > Personal), create it again, and finish the sign-in right away. See Connect to the PMA MCP Server from ChatGPT.
Tool Output and Data Issues
A query returns no rows for a platform you know is connected
Symptom: Asking the AI about a specific platform returns an empty result or the AI reports that it found no data, even though data for that platform is visible in your Hub.
Cause: One of three possibilities: (a) data has not yet synced for the requested date range, (b) the AI is querying an incorrect report type for that connector, or (c) a sync gap means no data exists for the window requested.
Solution: Ask your AI assistant to call pma_inspect_org_data for the connector. This diagnostic tool confirms which report types have data and what date ranges are populated. With that information, retry the query specifying the correct report type. If a sync gap is identified, initiate a manual backfill from Hub > Sources > Actions > Backfill data range for the affected date range.
An anomaly is flagged that appears to be missing data, not a performance change
Symptom: pma_detect_anomalies flags a day as an anomalous drop in spend or impressions, but investigation reveals no data was reported for that day at all; the drop is a sync gap, not a real performance change.
Cause: Before April 21, 2026, the anomaly detection tool treated zero-value days (caused by sync gaps) the same as genuine statistical outliers. This has been resolved.
Solution: As of April 21, 2026, pma_detect_anomalies returns a separate data_gaps field. Days with no data are now reported as gaps rather than as anomalies. If you are still seeing this behavior, verify that your AI client has fetched the latest tool description. If a data gap is confirmed in the data_gaps field, initiate a manual backfill via Hub > Sources > Actions > Backfill data range for the affected date range.
The same prompt returns different response shapes for different users or sessions
Symptom: Two team members ask the AI the same question and receive answers with different column names, different ordering, or different formatting.
Cause: The response-shape consistency limitation described in Data and Query Behavior above. Canonical response formats for high-frequency queries are still being standardized.
Solution: Until response shapes are standardized, include the desired output format in the prompt. For example: "List my data sources as a table with columns: Platform, Account Name, Has Data, Last Sync Date." This produces a predictable structure regardless of the underlying variation.
The AI makes many tool calls before answering, and responses are slow or expensive
Symptom: A question that seems simple triggers a long chain of tool calls before any answer is returned, which takes longer and uses more tokens than expected.
Cause: The cost consideration described in Cost and Performance Considerations above. Vague prompts require the AI to run discovery calls (pma_list_data_sources, pma_describe_platform, pma_inspect_org_data) before it can begin analysis, because it has to orient itself to your specific Hub configuration.
Solution: Specify the platform, metric, date range, and analytic intent up front. For a before-and-after example, see the Tips for Writing Effective Prompts section in the PMA MCP Tools Reference.
Wrong connector name
Symptom: A tool returns no results because the connector name in the prompt does not match a connected source.
Cause: Connector names must be passed exactly as returned by pma_list_data_sources.
Solution: Copy the connector name exactly from your connected sources list, or have the AI assistant call pma_list_data_sources first to retrieve the canonical name.
No data yet for a new connection
Symptom: A newly added data source returns empty results when queried.
Cause: The initial sync has not yet completed for the new connection.
Solution: Wait approximately one hour after connecting for the first sync to populate, then retry the query.
Occasional query timeout
Symptom: A tool call times out without returning data.
Cause: Transient backend slowness or load.
Solution: Retry the same query a few minutes later.
API Key and Custom Client Issues
403 error: "API tokens cannot be used for MCP connections"
Symptom: A request to the MCP server returns HTTP 403 with the message "API tokens cannot be used for MCP connections. Please connect via OAuth at https://hub.powermyanalytics.com to use MCP." This often shows up in a script or backend service that used to send an API key as a Bearer token.
Cause: The MCP server accepts only tokens created by the sign-in (OAuth) flow. API keys from MCP & API > REST API are for the REST API only, and the MCP server stopped accepting them on August 15, 2026. Generating a new API key won't help, because the MCP server rejects every key created on that page.
Solution:
- For an AI assistant: remove the API key from the connector settings and connect through the normal sign-in instead, using the setup guide for your client (linked under Related Articles).
- For a backend service, scheduled job, or other integration with no browser: call the Power My Analytics API with your API key instead. It reads the same warehoused data and shares the same per-plan limits. To pull data shaped like a Data Builder table, see Get Your Data Builder Table Data Into Your Own Code.
MCP client cannot establish a Streamable HTTP connection
Symptom: Your MCP client library fails to open a connection to the MCP endpoint, or the connection drops shortly after opening.
Cause: Your client library may default to the older Server-Sent Events transport, or your network may be closing long-lived HTTP responses.
Solution: Confirm your MCP client is using Streamable HTTP transport explicitly (not the older SSE transport from earlier MCP versions). If you are running behind a proxy or restrictive firewall, allow https://mcp.powermyanalytics.com and confirm long-running responses are not being closed prematurely.
Usage Limit Issues
Too many requests (HTTP 429) during a burst of tool calls
- Symptom: Tool calls start failing with "too many requests" or HTTP 429 partway through a task, often when an AI assistant runs many tool calls in quick succession or an investigation fans out across several platforms.
- Cause: Your organization exceeded its plan's requests-per-minute limit, which MCP and the PMA API share (see Cost and Performance Considerations above). This is a per-minute limit, not your billing-cycle quota.
- Solution: Wait about a minute and retry. Ask narrower questions (one platform, a defined date range, the metric you want) so your assistant makes fewer tool calls. Integrations that call the PMA API should slow their call cadence and retry with exponential backoff.
Monthly API/MCP quota exceeded (HTTP 429)
- Symptom: MCP tool calls and API requests fail with HTTP 429 and
monthly_quota_exceeded("Monthly API/MCP calls quota exceeded" or "Monthly API/MCP rows quota exceeded"), with the suggestion "Upgrade your plan or purchase a capacity pack to continue." On the MCP page, Total requests (MCP & API) or Total rows (MCP & API) is at or over its limit. - Cause: Your organization has used its shared MCP and API request or row quota for the billing cycle. This is a hard block, not a per-minute rate limit, so waiting a minute won't clear it.
- Solution: To continue right away on the Business, Grow, Pro, or Single (legacy) plan, open Billing > Add-ons > MCP & API > Increase Limits and buy a Capacity Pack ($10 one-time; +5,000 requests and +250,000 rows for the current billing cycle, up to 5 per cycle) or turn on Premium Unlock ($49.95/month; adds 100,000 requests and 5,000,000 rows per month to your plan's limits). See Increase Your MCP and API Limits with Capacity Packs or Premium Unlock. Otherwise, requests resume when the quota resets on your billing day. On the Custom plan, contact your account manager. To find what used the most, sort the Connected application table on the MCP page by Rows or Requests.
When to Contact Support
Submit a ticket if:
- You are experiencing an error or behavior not described in this article.
- The troubleshooting steps above did not resolve your issue.
When opening a ticket, include the AI client you are using (Claude Desktop, ChatGPT, or other), the prompt or action that triggered the issue, and any error message or unexpected output you received.