Skip to main content

Data Refresh by Platform Viewer

On this page you can get a high-level understanding of data refreshness at the platform level for each client across various data sources. This provides visibility into which platforms may be experiencing data delays without the granularity of individual accounts.

Below is a view that shows the most recent data date for each platform (data source) for all of Thrive's active clients. Data is automatically updated twice daily at 7:30 AM and 11:30 AM PST from our BigQuery data warehouse.

How to Use​

Understanding the Table​

The Data Refresh by Platform table displays the most recent date of available data for each platform (data source) for each client. This high-level view helps you quickly identify which platforms may be experiencing data delays across your clients without drilling into individual account details.

Table Columns​

  • Client: The client name
  • Platform: The advertising platform or data channel (e.g., Google Ads, Facebook Ads, Google Analytics)
  • Expected Lag: Whether this platform is expected to have data lag (Yes/No)
  • Expected Lag Days: The number of days of expected lag for this platform (shown as "N/A" if no lag is expected)
  • Most Recent Date: The latest date for which we have data for this platform across all accounts

Filtering the Data​

The table includes three interactive filters at the top:

Client Filter (Multi-Select)​

  • Select one or multiple clients to view only their platforms
  • Click "Select All" to choose all clients or "Clear All" to reset
  • Shows the number of clients selected (e.g., "3 selected")
  • Default: "All Clients" (shows all clients)

Platform Filter​

  • Select a specific platform (e.g., "Google Ads") to see that platform across all clients
  • Works interdependently with other filters
  • Default: "All" (shows all platforms)

Expected Lag Filter​

  • All: Shows all platforms regardless of expected lag status
  • Yes: Shows only platforms where data lag is expected
  • No: Shows only platforms where data should be current
  • Use this to focus on platforms that should have fresh data vs. those with acceptable delays

Smart Filtering: The three filters work together intelligently:

  • When you select clients, the Platform dropdown updates to show only platforms used by those clients
  • When you select a platform, the Client dropdown updates to show only clients using that platform
  • This makes it easy to answer questions like "Which clients use Google Ads?" or "What platforms does Client X use?"

Interpreting Data Freshness​

What "Fresh" Data Looks Like​

  • Ideal: Most Recent Date is yesterday or today (depending on time of day)
  • Expected with Lag: If "Expected Lag" is "Yes", data may be 2-3 days old and still considered normal

Warning Signs​

  • Stale Data (No Expected Lag): Most Recent Date is 3+ days old for platforms where Expected Lag = No

    • Possible causes: Platform-wide API issues, authentication problems, connector service down
    • Action: Check platform status, verify credentials, review API logs for this platform

  • Very Stale Data: Most Recent Date is 7+ days old regardless of Expected Lag setting

    • Possible causes: Critical platform failure, credentials expired, API access revoked
    • Action: Immediate investigation - verify platform status, check error logs, confirm API access

  • Missing Platform: An expected platform doesn't appear in the table for a client

    • Possible causes: Platform recently added but not yet ingested, platform disconnected, configuration error
    • Action: Verify platform configuration, check if platform is active in the source system

Common Use Cases​

1. Platform-Level Health Check​

Goal: Verify all platforms for a specific client are up-to-date

Steps:

  1. Select the client(s) from the Client dropdown
  2. Review all their platforms across all data sources
  3. Identify any platforms with stale dates
  4. Check "Expected Lag" column to determine if the delay is normal
  5. For unexpected delays, investigate platform-specific issues

2. Cross-Client Platform Investigation​

Goal: Check if a specific platform (e.g., Google Ads) is experiencing issues across multiple clients

Steps:

  1. Select the platform from the Platform dropdown
  2. Review all clients using that platform
  3. If multiple clients show stale data for the same platform, it indicates a platform-wide issue
  4. If only specific clients are affected, the issue is likely client-specific or configuration-related

3. Multi-Client Monitoring​

Goal: Monitor data freshness for a specific set of clients

Steps:

  1. Select multiple clients from the Client dropdown using the multi-select feature
  2. Review all platforms across the selected clients
  3. Quickly identify any platforms with stale data across your monitored clients
  4. This is useful for account managers monitoring specific client portfolios

4. Expected Lag Validation​

Goal: Verify that platforms marked with "Expected Lag" are within acceptable delay ranges

Steps:

  1. Select "Yes" from the Expected Lag filter
  2. Review all platforms with expected lag
  3. Confirm their Most Recent Dates are within the acceptable lag window
  4. If lag exceeds expectations, investigate further even for platforms with expected lag

Best Practices​

  1. Check Daily: Review this table each morning for a quick overview of platform health
  2. Understand Expected Lag: Familiarize yourself with which platforms are marked with Expected Lag and why
  3. Platform-Level Focus: This view provides a high-level overview - for granular account-level details, use the Data Refresh by Account view
  4. Multi-Client Selection: Use the multi-select client filter to monitor specific client portfolios efficiently
  5. Document Platform Issues: When you identify stale platform data, document the details in incident logs

Troubleshooting Guide​

Scenario 1: Single platform stale for one client, same platform fresh for other clients

  • Likely cause: Client-specific configuration issue or credentials

Scenario 2: All platforms from one client are stale

  • Likely cause: Client-level pipeline failure or configuration issue

Scenario 3: One platform is stale across multiple/all clients

  • Likely cause: Platform-wide API issue or connector service down

Scenario 4: Platforms with "Expected Lag = Yes" exceed normal lag window

  • Likely cause: Underlying data source delay beyond normal, or configuration needs updating

Scenario 5: New platform doesn't appear in table for a client

  • Likely cause: Platform not yet configured in pipeline, or initial load not complete