I have noticed some discrepancies when viewing reports in Community Analytics.
You will see different totals if you look at the displayed Unique Visitors number on the Traffic > Overview page...
... versus the sum of the number from downloading the Unique Visitor chart.
Both of these are from the same time period, why is the downloaded report showing a different number from what the dash board displays?
As I mentioned, same time frame, no filters, and a team member double checked my work. Not sure what is going on here, any insight is appreciated! Thanks!
When filtered down to the last 7 days, can there be an option to group the charts by "Week" instead of just "Day" or "Hour"?
I think the issue Rose and I are experiencing is because the Unique Visitor chart can only be grouped by "Day" so it's counting each unique visitor for that day of the week, not the FULL 7 days. That's creating duplicates.
We need to be able to schedule a report that gives us the correct number of unique visitors for the Last 7 Days, but can't find that data point as is.
Any help is appreciated! 🙂
I had actually had what sounds like the exact same question and submitted a case to understand further. Here is the explanation Khoros provided:
"The data for the number of visits is the same data. However, the three areas you highlighted in your screenshots that are labelled "Visits" are actually all measuring slightly different scenarios using the same master dataset.
The discrepancy between the Content total Visits at the top vs. the sum of the individual content categories is due to those numbers actually measuring different things.
One of my colleagues had a great analogy to help explain this: If you think of the community as a mall, then each visit to the mall would count as a visit in the Visits listed at the top of the page. Once you are in the mall, you may go to different stores. Each visit to a store (category) is marked in the visits column under the categories. You may go to multiple stores, but it would only count as one visit to the mall.
The best option for the total number of visits to your community will be the Visits number listed on the traffic overview page. This number will include visits to the community including the community landing page (extending the analogy, this could be seen as the sidewalk or front doors of the mall)."
That is a great explanation, thank you for taking the time to respond!
Here is an another open question, don't feel pressured to answer, but if you have any insight please chime in!
If we are utilizing a BI tool for our reporting and need to feed the correct number (In this case, that top dashboard number for visits) to the tool, then how would we accomplish this?
Here is some back story: Khoros doesn't integrate with our Business Intelligence tool so we use a Box connector instead. The Box files can be set up to update via email, which is what we are doing with the scheduled reports. However, we can't schedule a report that shows the Traffic Overview dashboard number (the number displayed at the top of the page), we have to use the Traffic Overview chart... which is not the metric we want to use.
Could you please provide details on which community is being checked and the time ranges under consideration, including the granularity of aggregation for the reports.
PageView: A page view is counted each time a browser requests a page.
Visits: A visit is one or more page views belong to one or more categories over time by a uniquely identified client (visitor) in a session over the specified time interval.
Unique Visitors: The number of unique visitors over the specified time interval. A unique visit may contain multiple visits by a user.
A pageview is a discrete event, but Visits and Unique visits are not discrete. A visit may contain multiple pageviews, these pages may belong to different categories/boards.
When you aggregated visits on category/board belongs to a visit, it is counted across multiple categories/boards. The same logic applies to Unique visitors. Adding up Visits and Unique visitors after category/board level aggregation like below is not a right metric. There is an overlap of a visit on different boards.
Let's take an example, a user with user-id:1234 visits (visit-id:visit111) to the community and activities are as follows.
For this user activity in a visit here is the metrics
In summary, metric counted a - 3 pageview - 1 visit - 1 Unique Visitor. ( 1 pageview attributed to community Home Page that may not belong to any category)
When aggregated on category/board level
Category-1/Board-1 Community. - 1 pageview - 1 visit - 1 Unique Visitor
If you do add up(like you did in CSV). 2 pageview - 2 visits - 2 Unique Visitor.
Adding up Visits and Unique visitors after category/board level aggregation is not the right metric.
if need you more information, we request to you file a support case. So that we can provide more information and data to support.
@SudhakaraS That implies that there is no accurate way to schedule a report for this metric, that we are limited by the platforms' capabilities when reporting on Unique Visitors. That is troubling.
@PrernaR It sounds like we will have to open a Support Case to determine why the community analytics tool has discrepancies between the dashboard display and the scheduled reports.
There is a product limitation right now which does not expose the summary metric in the Reports.
However here are two approaches you could consider,
(1)Please check once if you could use monthly granularity, then the sum up of individual rows(months) will be closer to the summary metric.
The current CSV Exports only export data at monthly, weekly, daily, hourly granularity.The higher the granularity the lesser the error.
Having said that, this is still not the most accurate mechanism to derive this metric.
(2) Other than that, we also expose the raw data via Bulk Data API which can be used to get accurate counts of the same metrics.
Please let us know if either of these two approaches would work.
Welcome to the Technology board!Curious about our platform? Looking to connect on social technology? You've come to the right place!