The method: how every impression is classified

One impression, one gate

Start at the DSP's own count. Every impression it billed walks the gates below and stops at the first one it fails. It gets one row, at that stage, with that reason.

1. 1

   Did the SSP serve it?

   no → no_ssp_record · yes → Served

2. 2

   Did our script run?

   no → tracker_never_fired · yes → Rendered

3. 3

   Could we measure it?

   no → xdomain_iframe · yes → Measurable

4. 4

   Did it meet the threshold?

   no → below_threshold · yes → Viewable

5. 5

   Did it pass the invalid-traffic checks?

   no → ivt_* · yes → Valid

An impression that clears every gate is valid — the thing you were actually buying. Add all the stages back up and you are at the DSP's count again.

When a partner cannot pass our ID

Some partners cannot put our impression ID in the tag. They send their own auction ID and we resolve it through the logs. That is a different quality of evidence, so it is reported as a different tier and never folded into an exact total.

Tier	How it was matched	Summable?
exact	Our impression ID, present on both sides.	Yes
partner_id	The partner's own auction ID, resolved through the logs. Still an exact join.	Yes, when labelled
fuzzy	Not built yet. Reserved for matches on time, campaign and creative within a window. Nothing is matched this way today, so no row carries this tier.	No
none	Evidence with no usable key at all.	No

A match rate that quietly includes guesses is worse than no match rate, because you would put it in a client report.

What the DSP never billed

Some impressions have SSP or beacon evidence but no DSP record at all. They are handled by a separate daily pass and carry stage = mismatch, so they are visible without ever entering a billed total.

• ssp_only — the SSP served it and the DSP has no record.
• tracker_only — a beacon for an impression neither log contains.
• unattributed — evidence with no usable key.

These are partner logging gaps, not wasted spend, and the reports treat them that way.

Industry standards

The definitions are the industry's, not ours. Where we follow a standard it is the one buyers already use; where we do not yet, it says so here rather than in a sales call.

Standard	What we apply
MRC viewability (display)	Applied. At least 50% of the ad's pixels in view for one continuous second, or 30% for ads of 242,500 pixels or more. Decided on our server from the evidence the tag sends, never by the tag.
MRC invalid traffic — general (GIVT)	Applied. Known bots and crawlers by their user agent; headless and automated browsers; pre-rendered and prefetched pages, which are not impressions under MRC.
MRC invalid traffic — sophisticated (SIVT)	Not yet. Traffic built to look human needs behavioural and network analysis we do not do. Our invalid-traffic figure is the general kind only.
IAB/ABC International Spiders & Bots List	Not yet. We match against our own list of crawler signatures. The licensed list is the industry's reference.
In-app (Open Measurement SDK) and video (VAST)	Not supported. Web display only.
MRC accreditation	Not accredited. We apply the standards above; accreditation is an independent audit that we apply them correctly, and we have not been through one.

What we will not do

• Fill gaps with estimates. An unmeasured impression is reported as unmeasured. Modelled numbers are indistinguishable from real ones the moment they enter a spreadsheet.
• Let the tag decide who can read anything. What you can see is decided by your credential, on every request. The billed reports take an impression's advertiser from the DSP log; the reports of what the tag measured alone have only the tag's own data-adv to go on, so a tag can decide which advertiser's report an impression is filed under — never who may read it.
• Present preliminary numbers as settled. Every response says which it is.
• Silently change history. A correction is a replay of a whole day, and reruns are idempotent by construction.