Mailjet

Mailjet exposes several object types through its API: campaigns, campaign drafts, statistical counters, contact lists, bounce logs, and click and open events. This integration focuses exclusively on the campaign object, which constitutes the primary unit of work for an email marketer and the only object type that supports reliable per-user attribution. Other objects, such as contact statistics or bounce logs at the recipient level, provide supplementary diagnostic data but cannot be attributed to a sender with sufficient precision for individual performance management. Five KPIs were retained, selected against three criteria: ability to attribute to a sender, resistance to gaming, and balance between leading and lagging indicators.

Campaign activity and engagement

Campaigns Sent counts the number of campaigns a sender has dispatched within the measurement period. It is the only leading indicator in this integration: it measures marketing execution volume, not marketing outcome. As a standalone metric, it is insufficient — a sender can dispatch a high volume of campaigns while achieving poor engagement or actively degrading the deliverability reputation of the sending domain. For this reason, Campaigns Sent is always read alongside the four ratio indicators. Its role is to provide the denominator context that makes those ratios interpretable: a 3% click-through rate on two campaigns and the same rate on twenty campaigns represent very different levels of evidence.

Click-Through Rate measures the average ratio of clicks to emails sent across a sender's campaigns. It is the primary engagement signal retained in this integration. Open rate was explicitly excluded: since Apple Mail Privacy Protection began pre-fetching email pixels in 2021, raw open counts are systematically inflated for a substantial share of the audience, making open rate an unreliable proxy for genuine interest. Click-through rate does not carry this distortion — a click requires an intentional action by the recipient and directly indicates that the content was sufficiently relevant to prompt a response. Among all available engagement metrics, it is the one most resistant to technical inflation.

Reading Campaigns Sent alongside Click-Through Rate surfaces a tension that is central to email marketing management. A sender who increases dispatch frequency without maintaining content quality will typically see click-through rate erode over time, as recipients become desensitized or disengage. Conversely, a sender with a high click-through rate on few campaigns may be under-exploiting the channel. The combination of the two indicators defines the productive envelope: enough campaigns to generate signal, with enough engagement to confirm the signal is real.

Deliverability health: three indicators of list and domain quality

Bounce Rate measures the average ratio of bounced messages to emails sent. A bounce indicates that an email could not be delivered: either the address no longer exists (hard bounce) or the recipient's mailbox is temporarily unavailable (soft bounce). The Bounce Rate is a direct indicator of list hygiene. Above certain thresholds, internet service providers begin throttling or blocking messages from the sending domain, which means a high bounce rate degrades deliverability for all campaigns sent from that domain, not only the campaigns targeting invalid addresses. This systemic effect makes Bounce Rate one of the most operationally urgent signals in the integration.

Unsubscribe Rate measures the average ratio of unsubscribes to emails sent. It is an early warning indicator of audience fatigue or content relevance decay. A rising unsubscribe rate often precedes a deterioration in click-through rate: recipients who no longer find value in the content will unsubscribe before the click rate visibly declines. Tracking both indicators together allows earlier detection of list health problems. Unsubscribe Rate also serves as a counterbalance to the gaming risk inherent in Campaigns Sent: a sender who artificially inflates dispatch frequency to show high activity will typically generate a proportional increase in unsubscribes, making the unsustainability of the behavior visible in the data.

Spam Complaint Rate measures the average ratio of recipients who marked the message as spam to total emails sent. It is the most critical deliverability indicator in this integration. Internet service providers apply automatic sanctions above a complaint rate of approximately 0.1%: messages from the sending domain are routed directly to spam folders, and at higher thresholds, the domain may be blacklisted entirely. Unlike Bounce Rate and Unsubscribe Rate, which measure passive failure states, a spam complaint represents an active negative signal from the recipient. The Spam Complaint Rate is non-gameable and unambiguous: there is no technical explanation for a high rate other than a mismatch between sender behavior and recipient expectations.

The three deliverability indicators form a coherent diagnostic set. Bounce Rate identifies list quality problems rooted in data hygiene. Unsubscribe Rate identifies relevance problems rooted in content or frequency. Spam Complaint Rate identifies consent problems rooted in acquisition practices or sending behavior. Each can be elevated while the others remain low, and the pattern of which indicators are elevated determines the appropriate corrective action. Reading them in combination provides a more complete diagnosis than any single indicator permits.

Scope and limits of the integration

Mailjet records campaign-level statistics but does not measure the business outcomes downstream of a click. A high click-through rate confirms that recipients engaged with the content; it does not confirm that those clicks generated conversions, revenue, or any other commercial outcome. The integration tracks email marketing performance as a marketing activity, not as a commercial result — the relationship between email engagement and commercial impact depends on factors outside the Mailjet data model.

Attribution in this integration depends on one structural assumption: each sender address registered in Mailjet must correspond to exactly one Human Bridge user. Where sender addresses are shared role accounts — a common pattern in organizations where multiple people send under a single marketing address — per-user attribution is not possible, and KPI values will reflect the aggregate output of the team rather than the contribution of any individual. The reliability of this integration is therefore conditional on the discipline with which sender address assignment is managed during onboarding and as the team changes.