4 ms·

For anyone else experiencing this problem what you can do is simply setup a filter in Google Analytics to only allow your domain to add traffic to your google a

For anyone else experiencing this problem what you can do is simply setup a filter in Google Analytics to only allow your domain to add traffic to your google analytics profile. Inside google analytics goto filter, create new filter, select custom filter, select include filter, Enter hostname in the filter field, enter your websitename\.com in the filter pattern box, Apply this filter to your profiles for websitename.com and you should be good to go

oniTony 13y ago
That's most of the way there, but if there's sufficient amount of combined traffic, ad hoc reports will start populating from sampled data instead of showing exact results. This might prevent getting any meaningful data about e.g. how A/B test is performing when filtered to a specific segment.
- ApolloRising 13y ago
  oniTony not sure what you mean exactly since this would just filter traffic to your own domain. This only works moving forward and usually is setup when a profile is created. Filters will not work retroactively with Google Analytics. Would be happy to discuss if you could clarify your statement regarding A/B testing.
  - oniTony 13y ago
    I went back to the Analytics Admin panel to double check, and it indeed looks like filters are applied to specific views. The tracking code collects data at the property level. From https://support.google.com/analytics/answer/2637192?hl=en https://support.google.com/analytics/answer/2637192?hl=en > If the number of visits/sessions to the property in the given date range exceeds 250K visits/sessions, GA will employ a sampling algorithm... > It is important to note that session sampling occurs at the property level, not the view level. So it sounds like reports will sample 250K from total traffic first, and apply the view filter after. This has the potential to be left with reports generated on too small of a sample.

oniTony 13y ago

That's most of the way there, but if there's sufficient amount of combined traffic, ad hoc reports will start populating from sampled data instead of showing exact results. This might prevent getting any meaningful data about e.g. how A/B test is performing when filtered to a specific segment.

ApolloRising 13y ago
oniTony not sure what you mean exactly since this would just filter traffic to your own domain. This only works moving forward and usually is setup when a profile is created. Filters will not work retroactively with Google Analytics. Would be happy to discuss if you could clarify your statement regarding A/B testing.
- oniTony 13y ago
  I went back to the Analytics Admin panel to double check, and it indeed looks like filters are applied to specific views. The tracking code collects data at the property level. From https://support.google.com/analytics/answer/2637192?hl=en https://support.google.com/analytics/answer/2637192?hl=en > If the number of visits/sessions to the property in the given date range exceeds 250K visits/sessions, GA will employ a sampling algorithm... > It is important to note that session sampling occurs at the property level, not the view level. So it sounds like reports will sample 250K from total traffic first, and apply the view filter after. This has the potential to be left with reports generated on too small of a sample.