The chart shows where a specific message copy was observed in a test mailbox. Matching folders on either side of a missing check may be joined by color. A continuous color does not prove continuous residence in that folder.
Method version 2026-10-01 · Updated October 1, 2026 · Published by Inbox Check / Live Direct Marketing
What is measured, and when?
The observation unit is one received message copy in one test mailbox. Each successful check records its folder and actual timestamp. The target interval is five minutes; delays and failed checks can leave gaps. Each copy’s 24-hour window starts at its first confirmed detection. A target schedule is not a record of completed checks. Times shown with UTC use the same time reference regardless of your local time zone.
Why do we fill some gaps with color?
For the same copy and mailbox, two confirmed observations of the same folder bound an interval that we display in that folder’s color. One or several missed checks may lie inside it; there is no fixed maximum gap length. Matching colors alone do not establish matching folders. Different folders or labels, and confirmed “copy not found” observations, break the interval.
This is a display assumption between observed endpoints. A message could move from Inbox to Spam and back between checks without us observing those changes. Two equal endpoints do not determine the probability of an unobserved move, its time, or the continuous time spent in a folder. A longer filled gap does not provide the same evidence as frequent actual measurements.
When does a gap stay gray?
Examples for one message copy in one mailbox. A dash means no measurement.| Recorded observations | Display |
|---|
| Inbox → — → Inbox | Continuous Inbox color |
| Spam → five gaps → Spam | Continuous Spam color |
| Inbox → — → Spam | Gray interval: transition time unknown |
| — → — → Inbox | Leading gaps stay gray |
| Inbox → five gaps → end of day | Trailing gaps stay gray; no right endpoint |
| Folder A → — → Folder B | Gray even if both folders use the same color |
| Inbox → copy not found → Inbox | The confirmed absence remains visible |
| Inbox → timeout → Inbox | Inbox color between endpoints; timeout retained in history |
We never fill future time or extend the final observation to the end of the day. A timeout means the check did not establish the folder; it does not prove that the message disappeared. A successful search that did not find the copy is a separate result and does not, by itself, prove deletion.
Does the filled color change the statistics?
No. Original observations, timestamps, gaps and errors remain unchanged. A filled interval adds zero completed checks and does not increase measured coverage. It is not used to calculate confirmed residence time or inbox percentage. The inbox percentage below uses actually received copies, not the length of colored bars. The tooltip and expandable mailbox details identify joined intervals.
How can the display rule be checked?
Consider this synthetic example, with all times in UTC: 10:00 Inbox; 10:05 no measurement; 10:10 no measurement; 10:15 Inbox; 10:20 no measurement; 10:25 Spam. The first interval is green from 10:00 to 10:15. The interval from 10:15 to 10:25 stays gray. There are three successful observations before and after rendering. Replacing the 10:15 observation with Spam makes the first interval gray. Ending the example at 10:10 leaves an unfilled trailing gap.
For each message copy and mailbox independently:
Order recorded observations by their actual time.
Keep only observations inside that copy's time window.
Keep confirmed absence and intermediate folder changes as boundaries.
For each pair of consecutive confirmed boundaries:
Same actual folder and labels: join by folder color.
Otherwise: leave the interval gray.
Keep every actual observation visible.
Do not extrapolate before the first or after the last observation.
Do not create or count inferred measurements.These examples describe the deterministic display rule, not a statistical validation of unobserved behavior. The individual-observation list in a report lets its reader compare actual timestamps with the chart. Historical email snapshots remain unchanged when the display rule is updated.
What does the daily summary mean?
The daily summary uses the last successful observation saved within each copy’s window, including checks between hourly chart points. It reports the folder and actual time, for example: “Last confirmed location: Inbox, checked 23 hours 35 minutes after observation began.” Creating the summary does not trigger a required extra mailbox check at exactly 24 hours. A later failure does not erase the earlier successful observation; neither proves the current folder.
The summary is scheduled after the observation window; mail transport and queue delays are separate from measurement time. Finishing the report does not certify an uninterrupted observation history. The 24-hour boundary is not an additional data point. If no confirmed location is available, the report says so.
Can the test predict a whole campaign?
The results apply to the observed copies in these test mailboxes at the recorded times. These mailboxes are not a random sample of all recipients. The results alone do not estimate a campaign-wide inbox rate, establish a cause of spam placement, or guarantee the outcome of a later message. An authentication pass and an Inbox placement are separate findings.
Technical references and scope
IMAP4rev2 (RFC 9051) specifies mailbox access and message operations. Authentication-Results (RFC 8601) defines reporting of authentication results. Microsoft Graph message documentation describes message properties available through that API. These sources explain protocol data; they do not validate our gap-filling convention. That convention and its limitations are specified above.
