Question 1

What email analytics does the Inbox OSS dashboard provide?

Accepted Answer

The dashboard provides **17 real-time analytics views** covering every dimension of email performance: send/delivered/opened/clicked/bounced/complained aggregates over any window, real-time event log queryable for 30 days, geo breakdown of opens and clicks by country derived from IP at engagement time, device and email-client splits parsed from User-Agent (Gmail, Outlook, Apple Mail, mobile vs. desktop), aggregate performance by recipient mailbox provider (Gmail, Outlook, Yahoo, corporate), engagement heatmaps showing when your audience opens, per-campaign deep-dives with funnel-from-send-to-conversion, A/B variant breakdowns, unsubscribe attribution, and revenue attribution that ties conversions back to specific campaigns and automations.

Question 2

How accurate is open rate tracking with iOS Mail Privacy Protection?

Accepted Answer

Apple's **iOS Mail Privacy Protection (MPP)** automatically pre-fetches tracking pixels for Apple Mail users, which inflates raw open counts for that segment. The Inbox OSS analytics dashboard handles MPP transparently in three ways: (1) we tag opens from Apple-Privacy-Relay IP ranges and surface them as "MPP-inflated" in the dashboard; (2) the headline open rate metric has an "exclude MPP" toggle that filters Apple Mail opens from privacy-relay ranges; (3) click-through-rate, conversion rate, and revenue attribution are unaffected because they require explicit subscriber action. This is more transparent than SendGrid and Mailchimp, both of which silently count MPP opens without flagging them.

Question 3

How does click tracking analytics work?

Accepted Answer

Click tracking rewrites every link in your sent emails to route through your **tracking domain** (e.g., click.yourcompany.com — you control the domain so links display as your brand). When a recipient clicks, the tracking endpoint logs the event with timestamp, IP, User-Agent, message ID, recipient, and the destination URL, then immediately redirects to the destination with a sub-50ms latency budget so the user does not notice. Click events are aggregated per-link per-campaign, deduplicated for unique-click metrics, and reported with click-map overlays showing which links in the template drove the most engagement. Click events also fire your webhook stream in real time for application-side processing.

Question 4

Can I track email ROI and attribute revenue to specific campaigns?

Accepted Answer

Yes. Revenue attribution works through two mechanisms: a **tracking pixel** on order-confirmation pages that pings our attribution endpoint with the transaction value and order_id (deduplicated to prevent double-counting on page refresh), and a **server-side webhook** from your backend that posts the conversion with the campaign attribution token preserved in the original message link. The platform stitches conversions back to the specific campaign, variant, and subscriber, and surfaces revenue-per-send, revenue-per-recipient, and total attributed revenue in the analytics dashboard. Attribution windows are configurable (default 7-day click, 1-day view); the attribution model can be switched between last-touch, first-touch, and linear depending on your marketing stack convention.

Question 5

How long are email events retained in the analytics dashboard?

Accepted Answer

The **real-time event log retains every event for 30 days** with full sub-second query latency over your entire account history within that window. Aggregated metrics (daily/weekly/monthly rollups for opens, clicks, bounces, etc.) are retained indefinitely so you can chart year-over-year performance. For longer raw-event retention, the platform exports the event stream to your S3 bucket, BigQuery, Snowflake, or any HTTPS endpoint via our webhook firehose — most customers pipe it into their data warehouse for joining with product analytics and BI tooling. Webhook delivery is at-least-once with Ed25519 signatures and exponential-backoff retries.

Question 6

How does mailbox-provider performance reporting help my deliverability?

Accepted Answer

Mailbox-provider reporting aggregates every metric by recipient domain — Gmail, Outlook, Yahoo, Apple, and the long tail of corporate and ISP domains. This is the single most important deliverability dashboard because **inbox placement varies wildly by mailbox provider**: you can have great Gmail engagement and terrible Outlook engagement simultaneously, and the headline open rate hides the provider-specific failure. The dashboard surfaces per-provider delivered-rate, soft-bounce rate, hard-bounce rate, complaint rate, and open rate, with multi-week trend lines so you can spot reputation degradation at one provider before it spreads. Customers fixing per-provider reputation issues typically see 5-15 percent inbox-placement lift within 14 days.

Question 7

Can I segment analytics by category, tag, or custom argument?

Accepted Answer

Yes. Every event carries the **categories** and **custom_args** stamped on the original message, so every metric in the dashboard can be filtered by any category or custom-arg combination. Practical example: tag every onboarding-flow message with category=onboarding and custom_args.cohort=2026-03, then filter the dashboard to "category=onboarding AND cohort=2026-03" to see open and click metrics for just that cohort. Tag combinations support AND/OR/NOT logic. This is meaningfully more flexible than SendGrid Stats, which only filters by categories (not custom_args) and only supports up to 10 categories per message.

Question 8

How does engagement heatmap reporting work?

Accepted Answer

The engagement heatmap surfaces **when your audience actually opens** — day of week and hour of day — broken down per segment, per category, or globally across the account. Data is computed from open events over the last 90 days (with MPP-affected opens optionally excluded). The heatmap is the input to send-time optimization: a peak at 9am Tuesday means scheduling sends for that window probably outperforms 2am Sunday. For per-recipient send-time optimization, the same engagement-pattern data is computed per subscriber and fed into our ML model to pick the optimal send moment for each individual.

Question 9

How do unsubscribes and complaints show up in the dashboard?

Accepted Answer

Unsubscribes and spam complaints have dedicated views with both rate trends and per-campaign attribution. The **unsubscribe attribution view** shows which campaigns drove the most opt-outs (often surfacing a problematic subject line or send frequency); the **complaint attribution view** does the same for spam complaints from feedback loops at Gmail, Outlook, Yahoo, and AOL. Spam-complaint rate above 0.1 percent is the industry warning threshold and the dashboard alerts you when a campaign or sender segment exceeds it. This is more direct than SendGrid, which shows aggregate complaint rate without per-campaign breakdown unless you pay for the Premium Stats add-on.

Question 10

Can I export raw event data to my data warehouse?

Accepted Answer

Yes. The platform supports three export modes: (1) **webhook firehose** that streams every event to your HTTPS endpoint in real time with Ed25519 signatures, exponential-backoff retries, and a replay window; (2) **scheduled CSV/Parquet exports** delivered to your S3 bucket on a daily or hourly cadence with full event payloads; (3) **direct integrations** for BigQuery, Snowflake, Redshift, and Databricks via Fivetran and Airbyte connectors. Most customers run the firehose for real-time event-driven workflows plus scheduled Parquet exports for warehouse loading. There is no separate charge for either — both are included in every paid plan.

17 dashboards. One source of truth.

Summary dashboard

Real-time event log

Geo breakdown

Devices & clients

Mailbox providers

Engagement heatmap

Per-campaign deep-dive

Revenue attribution

Frequently asked questions about email analytics

Know what worked. Repeat it.