[Publish] Data discrepancies between Buffer Publish and Twitter Analytics
There are a couple of reasons you might see a discrepancy between the data in Buffer Publish and the data in Twitter Analytics.
If you’re on the Pro Plan, click metrics shown in the Posts Report will be pulled from Bit.ly. It’s important to note that we are only able to track clicks on shortened links. If link shortening is turned off, click metrics will not be available. Bit.ly counts the total number of times a link has been clicked. This means that if the exact same link has been shared more than once, or to multiple social accounts, the number you'll see will include all of those clicks. Twitter on the other hand, will simply count the number of clicks on the one specific tweet. This means you may notice a discrepancy between click metrics in Buffer Publish compared to Twitter Analytics. Learn more about Bit.ly link tracking, and how you can get more information on your clicks, here.
If you’re on the Business Plan, click metrics in all reports will be pulled from Gnip, which is Twitter’s enterprise data service. All of the data we get from Gnip is on a per-tweet basis and we're only able to check for new engagements a limited number of times on each tweet. In practice, we're able to get data for seven days following a tweet being published. That means if you sent a tweet on Monday, we’ll keep checking it for new engagements and impressions until the following Sunday. After that, we can no longer re-check the tweet, and from then onward, its metrics will remain the same within Buffer Publish. Twitter, on the other hand, will continue to track data perpetually, meaning the data will continue to be updated as you receive more engagements after seven days.
Additionally, Buffer Publish does not currently track data on retweets or replies in the Overview Report or the Analysis Report, which is why you might notice the number of tweets may be lower than what you see in Twitter Analytics. The reason we don’t include retweets or replies is that they can skew the most useful insights that help measure and report performance on Twitter. Without replies and retweets, we’re able to more accurately measure and report your true performance on Twitter. For example, some brands use Twitter for customer service and therefore have a tremendous number of replies in a given time period. By their nature, replies have very low overall engagement and could make it otherwise appear like your total engagement is very low, when in reality your non-reply tweets are performing very well.