Improving WhatsApp delivery reliability
WhatsApp had become a primary channel for several enterprise accounts, with template-based messaging governed by strict provider rules.
PROBLEM
Delivery rates were inconsistent across accounts, and the failures surfaced to the customer as 'the message never arrived' with no further detail.
HYPOTHESIS
The failures were not random. They clustered around template state, opt-in status and rate limiting rather than around infrastructure.
RESEARCH
Segmented delivery outcomes by template, by account, and by provider response code, then compared healthy accounts against struggling ones to isolate the differing variable.
DATA USED
Provider webhook callbacks · template approval state · consent and opt-in tables · throughput limits
DECISION MAKING
Separated causes we controlled from causes the provider controlled, and only escalated the former as product work. Being explicit about that boundary kept the conversation with engineering credible.
SOLUTION
Clearer failure surfacing back to the campaign UI, template-state checks ahead of send, and documented throughput guidance for high-volume accounts.
BUSINESS OUTCOME
Customers could self-diagnose the common cases, and the recurring escalations tied to template state largely disappeared.
REFLECTION
Delivery reliability is a trust metric, not a technical one. The fix that mattered most was making the failure legible to the customer.
SQL INVESTIGATION
SELECT template_id,
provider_code,
count(*) AS n,
round(100.0 * sum(delivered) / count(*), 2) AS delivery_pct
FROM whatsapp_events
GROUP BY template_id, provider_code
HAVING count(*) > 500
ORDER BY delivery_pct ASC;