Let us start off by sincerely apologizing for the series of statistics-related incidents in the past days 💔
This should not happen, and we truly apologize for any negative impact these incidents may have on your business's ability to efficiently evaluate your catalog campaign's effectiveness.
We acknowledge the significance of precise data and how it can influence your business decisions.
Therefore, we are implementing measures to prevent such occurrences from happening again.
Please be assured that the iPaper platform's integrity and stability remain our top priority.
We aim to do – and be – better than this!
At this time, we hope for, and appreciate, your understanding.
If you have additional questions, our team is available to assist you in any way we can.
Please read on for more information about this Incident.
While working to optimize our time tracking accuracy, we introduced an edge case that has negatively impacted a group of customers. In certain cases, the registration would end up counting more page views than it should have, resulting in inflated page views and time tracking.
Error was introduced on March 21st, 2023 at 10:10 AM CET
Error was resolved on April 4th, 2023 at 01:32 AM CEST
See related incident here:
No tracking of heatmap hotspots
This challenge has not impacted every account/customer.
But – unfortunately, we have identified a group of customers which have been affected negatively.
These customers have experienced massively inflated page views and time spent on certain Flipbooks, often with more average page views than total pages in the Flipbook.
We will reach out to the identified accounts directly, with a list of the affected flipbooks and options to proceed.
There is no great solution, as we are unable to restore the data.
We can offer one of two options:
Reach out to us, if you would like to have the data reset (option 2).
This was due to a missed edge case in our handling of statistics, while we do our very best to verify code changes that go into our platform, we do make mistakes. With that said, we should have been able to catch this a lot faster, and we will be strengthening our ability to detect statistics anomalies, and look into ways we can potentially improve our ability to restore data in case of mistakes like this in the future.