Every metric the InboxAlly MCP server reports, with its unit and definition. Each answer from the server also carries the definitions of the metrics in it.
For a fuller explanation of ia_score_pct, see What is my IA Score and what does it mean?. For reputation_level, see What is my IA Reputation and how does it affect deliverability?.
| Metric | Name | Unit | Definition |
|---|---|---|---|
detected_emails | Detected emails | Count | Seed copies InboxAlly detected in its seed mailboxes: inbox, promotions, social and spam. Not emails sent through your ESP. |
detected_emails_30d | Detected emails, last 30 days | Count | Seed copies InboxAlly detected for this sender over the last 30 days: inbox, promotions, social and spam. Not emails sent through your ESP. |
share_of_detected_pct | Share of detected emails | Percent | This row’s detected emails as a percentage of every matching group’s detected emails in the period, not just the rows shown. |
primary_inbox_pct | Primary inbox | Percent | Detected emails that landed in the primary inbox, as a percentage (0-100) of detected_emails. |
inbox_incl_promotions_pct | Inbox including promotions | Percent | Detected emails that avoided spam (primary inbox, promotions and social), as a percentage (0-100) of detected_emails. |
promotions_pct | Promotions | Percent | Detected emails in promotions or social tabs, as a percentage (0-100) of detected_emails. |
spam_pct | Spam | Percent | Detected emails that landed in spam, as a percentage (0-100) of detected_emails. |
delta_pts | Change (points) | Percentage points | The change in a percentage between the previous period and this one, in percentage points. |
delta_rel_pct | Change (relative) | Percent | The change between the previous period and this one, relative to the previous value, in percent. |
seed_open_pct | Seed opens | Percent | The share of this sender’s seed copies that seeds open, 0-100. A setting, not a measurement. |
opener_scroll_pct | Scroll (of opened) | Percent | Of the seed copies that are opened, the share that seeds scroll through, 0-100. |
opener_spam_to_inbox_pct | Spam to inbox (of opened in spam) | Percent | When a seed opens one of this sender’s copies that sits in spam, how often it moves it to the inbox, 0-100: a share of the opened copies in spam. Copies that are never opened are never moved. |
opener_promotions_to_inbox_pct | Promotions to inbox (of opened in promotions) | Percent | When a seed opens one of this sender’s copies that sits in promotions, how often it moves it to the primary inbox, 0-100: a share of the opened copies in promotions. Copies that are never opened are never moved. |
opener_mark_important_pct | Mark important (of opened) | Percent | Of the seed copies that are opened, the share that seeds mark as important, 0-100. |
opener_reply_pct | Reply (of opened) | Percent | Of the seed copies that are opened, the share that seeds reply to, 0-100. |
opener_click_link_pct | Click link (of opened) | Percent | Of the seed copies that are opened, the share in which seeds click the configured link, 0-100. |
ia_score_pct | IA Score | Percent | InboxAlly’s estimate of the chance that a sender’s next seed copy avoids spam (lands in the inbox, promotions or social), 0-100, from a Bayesian model of recent seed results that weighs recent days more. The overall score blends providers and leans toward the weakest, so it can be lower than every provider’s score. |
weakest_provider_ia_score_pct | Weakest provider’s IA Score | Percent | The IA Score of the provider where this sender scores lowest, 0-100. |
ia_score_ci_low_pct | IA Score, lower bound | Percent | The lower end of the 95% credible interval around a provider’s IA Score, 0-100. |
ia_score_ci_high_pct | IA Score, upper bound | Percent | The upper end of the 95% credible interval around a provider’s IA Score, 0-100. |
evidence_weight | Evidence weight | Count | How much evidence a provider’s IA Score rests on: the model’s decayed pseudo-counts, prior included. Higher means a firmer score. Not a count of seed copies. |
latest_day_seed_copies | Seed copies (latest day) | Count | Seed copies InboxAlly detected for this sender at this provider on the latest scored day. |
reputation_level | IA Reputation | Level | InboxAlly’s IA Reputation: its own machine-learning prediction of the sending domain’s reputation, HIGH, MEDIUM, LOW or BAD. Google Postmaster Tools data is one optional set of its signals; it isn’t a mailbox provider’s own rating. |
reputation_confidence_pct | Prediction confidence | Percent | The model’s confidence in its predicted reputation level, 0-100. |
forensic_samples | Forensic samples | Count | Seed copies whose headers InboxAlly kept for authentication forensics, read in this answer. A sample of the seeds’ mail, not a rate for all of your mail. |
forensic_samples_matching | Matching forensic samples | Count | Every kept seed copy matching the request in the window. More than forensic_samples when the list was cut off. |
auth_result | Authentication result | Level | The receiving provider’s own verdict for SPF, DKIM or DMARC, as its Authentication-Results header states it: pass, fail, softfail, neutral, none, temperror, permerror or policy. other for any word outside those; not_reported when the receiver didn’t state that check; unreadable when its header couldn’t be read. |
auth_aligned | Aligned | Level | Whether the SPF or DKIM domain shares the From domain’s organizational domain (DMARC relaxed alignment). DMARC passes only on an aligned SPF or DKIM pass. unknown when either domain can’t be placed on the Public Suffix List. |
dmarc_policy | DMARC policy | Level | The policy the From domain publishes (p=: none, quarantine or reject), as the receiving provider stated it. other for any other value; not_stated when the receiver didn’t say; Microsoft never does. |
dmarc_disposition | DMARC disposition | Level | What the receiving provider says it did under DMARC (dis=, or Microsoft’s action=): none, quarantine, reject or oreject (Microsoft overriding a reject), other for any other value, not_stated when the receiver didn’t say. A DMARC fail with quarantine is a DMARC-driven quarantine; with none, it’s monitoring only. |
published_record_status | Published record status | Level | A record as the domain publishes it in DNS, when InboxAlly last looked: present; missing (looked, and none is published); multiple (more than one, which receivers treat as none; records lists each); invalid; null_mx (the domain accepts no mail); for DMARC, reporting (a rua address sends aggregate reports to InboxAlly) or not_reporting_to_inboxally; lookup_failed (DNS didn’t answer, so the record is unknown, never missing). DKIM isn’t checked this way, since no selector is known for a domain: the seed results’ DKIM verdicts are the evidence. |
spf_lookups | SPF DNS lookups | Count | The DNS lookups a receiver makes to evaluate the SPF record for a sender it matches nothing for, counted per RFC 7208 through includes and redirects; over 10 is a permanent error. void counts lookups that find nothing for every sender, over 2 an error; void_by_family counts what an IPv4 or an IPv6 sender meets. When complete is false, a nested lookup didn’t answer or counting stopped at 30, so the counts are a minimum. |
published_issue | Published record issue | Level | A problem InboxAlly found in a published record: a code, a severity (info, warn or error) and a message in InboxAlly’s words. Word it from the message, not from the record’s other fields: plus_all can fire while spf.all is -all or ~all, through an include that ends in +all. |
needs_attention | Needs attention | Level | Whether a sending domain’s published records need a look: true when SPF or DMARC isn’t published, a record is published more than once or is invalid, or a record has an issue of severity error or warn, each named in reasons; false when every record was looked up and none of that applies; null when it can’t be told, because the domain wasn’t checked or part of its answer is unknown (a lookup that didn’t answer). An unpublished BIMI or MX record, and DMARC reports that don’t go to InboxAlly, don’t count. DKIM isn’t checked this way. |
creative_change_pts | Creative change | Percentage points | The change in a creative’s primary_inbox_pct against its comparison point, in points, weighted across providers by where the creative was seen. The comparison point is the creative’s previous week (compared_with previous_week). null for a single send (compared_with sender_other_mail), which has no change over time. |
sender_change_pts | Sender change | Percentage points | The same change for the creative’s sender as a whole, against the same comparison point, at the same providers and with the same weights, in points. null for a single send. |
divergence_pts | Divergence from sender | Percentage points | How far the creative is from its own sender, in points: creative_change_pts minus sender_change_pts, or, for a single send (compared_with sender_other_mail), the send’s primary_inbox_pct minus the sender’s other mail on the same days and providers, weighted by the send’s seed copies. |
coverage_pct | Comparison coverage | Percent | The share of the creative’s seed copies in the compared week (for a single send, over the whole send) at providers with enough evidence to compare, 0-100. |
week_change_pts | Week-over-week change | Percentage points | The change in a creative’s primary_inbox_pct from one week to the next, in points. |
seed_limit | Seed mailboxes | Count | How many seed mailboxes the plan gives the workspace: the size of its seed list, before any bonus seeds (additional_seed_limit). A standing count, not a daily figure, and not a limit on the email sent to your own contacts. |
additional_seed_limit | Bonus seed mailboxes | Count | Seed mailboxes InboxAlly adds to your workspace on top of its plan’s seed_limit: the seed list holds seed_limit plus these. |
seed_emails_per_day | Seed emails per day | Count | The most emails a day the plan expects InboxAlly’s seed mailboxes to receive, counted across all of them: send the seeds no more than this each day. It counts mail to the seeds, not the email sent to your own contacts. null when the account has no plan. |
sender_profile_limit | Sender profiles | Count | The most sender profiles (From addresses InboxAlly monitors) the plan allows. |
seed_sends_per_day | Seed sends per day | Count | The daily sends to InboxAlly’s seed mailboxes that the warmup planner recommends once the seeds have ramped up. A plan, not a measurement. |
seed_share_pct | Seed share of volume | Percent | seed_sends_per_day as a percentage (0-100) of the daily volume the plan targets. |
test_inbox_pct | Test inbox rate | Percent | Seeds of this placement test where the campaign landed in the inbox, as a percentage (0-100) of the seeds checked: inbox, spam or missing. Seeds that couldn’t be checked, or are still pending, are left out. |
test_spam_pct | Test spam rate | Percent | Seeds of this placement test where the campaign landed in spam or junk, as a percentage (0-100) of the seeds checked. |
test_missing_pct | Test missing rate | Percent | Seeds of this placement test the campaign never reached in a finished measurement, as a percentage (0-100) of the seeds checked. Missing isn’t spam. |
seed_sends | Seed sends | Count | Sends to InboxAlly’s seed mailboxes planned for one day of a warmup plan. |
list_sends | List sends | Count | Sends to your own contacts planned for one day of a warmup plan. |