Analytics and Tracking

(not set)

Also called not set, missing dimension value

The placeholder a report prints when a dimension has no collected value for that row, rather than guessing one.

Quick facts: (not set)

Category
Analytics and Tracking
Also called
not set, missing dimension value
Level
Beginner
Affects
Attribution accuracy, campaign reporting, landing page analysis, segment reliability
Where to see it
GA4 reports and explorations, Looker Studio, Search Console, UTM builder
In this article4
  1. What (not set) means
  2. Why (not set) matters
  3. Common mistakes with (not set)
  4. How to act on it

What (not set) means

Every row in a report is a combination of dimensions — source, campaign, page, country, device. When the platform has no value for one of those dimensions on that row, it does not leave the cell blank or guess. It prints (not set), which reads as: something was measured here, but this particular attribute was never supplied.

The cause depends entirely on which dimension is showing it, and that is the first question to ask. Under campaign or content dimensions it usually means the link had no campaign tags, so there was nothing to record. Under landing page it often means the session’s first hit was missed, commonly because a tag loaded late or a redirect swallowed it. Under a custom dimension it means the event fired without that parameter attached. Under a geographic or device dimension it means the platform could not resolve the value from the request. One label, several unrelated stories.

Why (not set) matters

It is the honest part of the report. A platform that quietly assigned those sessions to the nearest plausible bucket would be far more damaging, because the error would hide inside a channel that looks real. Instead the gap is labelled, and a large one is a direct instruction to go and fix the collection.

The size is the signal. A small amount is normal in any account and not worth chasing. A large share sitting in one dimension usually means something structural: an untagged newsletter, an email platform stripping parameters, a campaign someone launched without UTM parameters, or a redirect chain that loses the query string before analytics ever sees it.

Common mistakes with (not set)

The most common is filtering it out. Removing the rows makes the chart tidy and makes the account look better attributed than it is, which is the opposite of useful. If a channel’s real contribution is hidden inside that bucket, deleting the bucket deletes the evidence.

The second is treating every occurrence as the same fault. (not set) under campaign is a tagging job for the marketing team; (not set) under landing page is a technical job involving tags and redirects. Applying the wrong fix wastes a week. A third is confusing it with (other), which is a bundle of values that did exist but were grouped together, not a missing value at all.

How to act on it

Start by finding out which dimension is producing it and how large its share is compared with the rest of the report. Then follow the cause rather than the symptom: tag every outbound link you control — email, social bios, QR codes on printed material, WhatsApp broadcasts — using one consistent naming convention, and use a builder so the values do not drift between campaigns.

On the technical side, check that the analytics tag fires on the first page of the session and that redirects preserve the query string, since both are frequent causes of missing landing pages. If the share stays high after that, a proper review of the analytics and tracking setup will usually find a single structural cause behind most of it.

Do and do not

Do

  • Check which dimension is producing it before diagnosing
  • Tag every link you control with consistent campaign parameters
  • Verify the tag fires on the session's first page

Do not

  • Filter the rows out to make a report look tidy
  • Assume every instance has the same cause
  • Confuse it with (other), which bundles values that did exist

Questions people ask about this

Is (not set) a tracking error?

Not always. It simply means a dimension had no value for that row, and some of that is unavoidable in any account. It becomes a problem when one dimension carries a large share of it, which normally points to untagged campaigns, a tag that loads too late, or redirects that drop the query string before analytics can read it.

Should I filter (not set) out of my reports?

No. Removing it makes the report look better attributed than it really is and hides the traffic you most need to explain. Leave it visible, note how large it is, and treat a growing share as a job to fix at the source. Tidying the chart does not tidy the underlying data.

What is the difference between (not set) and (other)?

(not set) means no value was collected for that dimension on that row. (other) means values were collected, but the report had more distinct values than it could list, so the remainder were bundled into one line. The first is a collection gap you can fix with tagging or tag timing; the second is a reporting limit you work around with different dimensions.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.