WhatsApp template analytics and comparison
Use an Application API key in x-api-key. Find the channel and internal template UUIDs through the template listing endpoint. Provider template IDs, app credentials and WABA IDs are resolved by Hub and are not API inputs. Both resources support connected WhatsApp channels; sandbox analytics are unavailable.
Template analytics
curl --get 'https://app.chatt2.me/v1/communication/channel/whatsapp/templates/CHANNEL_UUID/TEMPLATE_UUID/analytics' \
--header 'x-api-key: YOUR_APPLICATION_API_KEY' \
--data-urlencode 'start=2026-11-01' \
--data-urlencode 'end=2026-11-05'
start and end are inclusive UTC dates. Request at most 30 days, with the start within the preceding 90 days and no future dates. Today's values can be partial. Provider data may arrive late.
The result includes sent/delivered/read totals, daily series and typed button clicks. Missing values are null, not zero; missing days remain gaps. Unique button clicks are daily counts, not unique recipients across the whole period. Costs, MM Lite and offsite conversion analytics are not included in this resource.
Results share a five-minute cache. fetchedAt is Hub retrieval time, not provider freshness. The conservative provider budget is five reads per minute shared across the integration, including MCP. Cached reads do not spend provider quota. Responses use Cache-Control: no-store to prevent browser/shared HTTP caching of tenant data.
Typical errors: TEMPLATE_ANALYTICS_INVALID_PERIOD (400), TEMPLATE_ANALYTICS_NOT_ENABLED (409), TEMPLATE_ANALYTICS_UNSUPPORTED_CHANNEL (409), TEMPLATE_ANALYTICS_PROVIDER_ACCESS or TEMPLATE_ANALYTICS_INVALID_RESPONSE (502), and TEMPLATE_ANALYTICS_UNAVAILABLE (503). A template or channel outside the key's application returns 404. Enablement is managed by the connected application; a GET does not enable analytics.
Compare two templates
curl --get 'https://app.chatt2.me/v1/communication/channel/whatsapp/templates/CHANNEL_UUID/BASE_TEMPLATE_UUID/comparison' \
--header 'x-api-key: YOUR_APPLICATION_API_KEY' \
--data-urlencode 'compareTemplateId=OTHER_TEMPLATE_UUID' \
--data-urlencode 'periodDays=30' \
--data-urlencode 'endDate=2026-11-05'
Both internal template IDs must be distinct, approved and owned by the same channel. Each template needs at least 1,000 messages sent in the selected period. Use periodDays 7, 30, 60 or 90; endDate is the final included date, earlier than today UTC. Hub derives exact UTC boundaries in epoch seconds; clients do not send timestamps. The interface defaults to the latest complete 30 days.
The response includes:
| Field | Meaning |
|---|---|
messageSends | Map of internal template UUID to count, or null if absent |
topBlockReasons | Map of internal template UUID to the provider reason, or null |
blockRateOrder | Relative ordering supplied by the provider, mapped to internal UUIDs |
blockRateOrderMeaning | Currently unconfirmed; the provider does not document the direction clearly |
startDate, endDate, timezone | Actual inclusive dates and UTC |
fetchedAt, cached | Retrieval metadata |
Do not infer a winner or a numerical block rate from relative ordering. Greater sends do not mean better quality. UNKNOWN_BLOCK_REASON means no reported reason. Comparing different categories, languages or audiences does not create a controlled A/B experiment or statistical confidence.
Comparisons use a one-hour cache and a separate conservative budget of 10 provider calls per hour shared across the integration. REST and operational MCP use the same cache and quota. Selecting a different template in the UI does not run a comparison until Compare is selected. Reversing the base/comparator roles is a distinct query.
Errors and manual retry
| Code | HTTP | Recovery |
|---|---|---|
TEMPLATE_COMPARISON_INVALID_PERIOD | 400 | Choose a supported complete UTC period |
TEMPLATE_COMPARISON_SAME_TEMPLATE | 400 | Choose another template |
TEMPLATE_COMPARISON_INSUFFICIENT_VOLUME | 409 | Try a longer supported window; both templates need 1,000 sends |
TEMPLATE_COMPARISON_NOT_APPROVED | 409 | Wait for approval and refresh template metadata |
TEMPLATE_COMPARISON_NOT_OWNER | 409 | Verify both templates belong to this connected channel |
TEMPLATE_COMPARISON_UNSUPPORTED_CHANNEL | 409 | Use a connected WhatsApp channel |
TEMPLATE_COMPARISON_PROVIDER_ACCESS | 502 | Ask the application administrator to check the provider connection |
TEMPLATE_COMPARISON_INVALID_RESPONSE | 502 | Provider response could not be safely interpreted |
TEMPLATE_COMPARISON_UNAVAILABLE | 503 | Retry later |
For both reads, rate limiting returns HTTP 429, Retry-After and a string retryAfter in the error body. Respect the delay and retry manually. Cached results may be delayed too; neither endpoint promises real-time data.
Operational MCP
Use hub_get_whatsapp_template_analytics and hub_compare_whatsapp_templates with an authorized organizationContextId, applicationId, applicationChannelId and internal template IDs. Analytics also takes start/end; comparison takes templateId, compareTemplateId, periodDays and endDate. The services, scope checks, result semantics and provider quotas are shared with REST. Neither read sends a message or a conversion.
See MCP actions for browser OAuth and team selection, and Purchase conversions for CTWA event submission.