Why your data in Buffer Analyze might differ from your native analytics

This guide aims to outline the reasons why your data in Buffer Analyze may differ from the data you see in the native analytics for Facebook, Instagram, and Twitter.

This article contains the following sections:

  1. Applicable to all
  2. Facebook
  3. Instagram
  4. Twitter
  5. Other articles that might be helpful

Applicable to all

  1. When you first connect a social channel, we will backfill a certain amount of data, as described in the table below. More comprehensive details on this are available in the channel specific sections below.

    Channel Backfill information
    Facebook Six months of data is backfilled.
    Instagram Six months of data is backfilled, not including stories data.
    Twitter 28 days of data is backfilled.
  2. Data is not collected in real time, rather it is collected in batches from each social network's API each day. In some cases, it can take up to 24 hours for new post data to be populated.

  3. If your social account is disconnected, your latest data won’t be populating, so you’ll want to get that connection reestablished, following the steps here.

Facebook

  1. Due to API limitations, we are only able to pull data into Analyze for Facebook accounts that have 100 or more followers. Whilst it's technically possible to connect Facebook accounts that have fewer than 100 followers, no data will be available.

  2. Six months of Facebook data is backfilled when you connect your account to Analyze. This could result in data for previous periods appearing lower than in Facebook Insights. Let’s say your Facebook account was connected on October 1, 2019. We’d go back and fetch data starting from April 1 2019. In January 2020, if you looked at your data in Analyze for the whole year of 2019, it wouldn’t match what you see in Facebook Insights for 2019 because we are only capturing nine months of data (the six months prior to connecting the account, which is the data we’ve backfilled, and the three months after connecting the account).

  3. Audience information for Facebook will take up to 24 hours to populate.

  4. Data on individual Facebook posts will not be updated indefinitely. After a full calendar month has passed, we will no longer update the metrics for that post.

Instagram

  1. Due to API limitations, we are only able to pull Instagram data into Analyze for business accounts that have 100 or more followers. Whilst it's technically possible to connect business accounts that have fewer than 100 followers, no data will be available.

  2. Six months of Instagram data is backfilled when you connect your account to Analyze.

  3. Whilst we're able to backfill six months of Instagram post metrics when you connect your account, unfortunately, we’re not able to backfill Instagram stories. We are only able to start collecting that data from when your Instagram account is connected. Additionally, if your Instagram account becomes disconnected from Analyze, any stories you post during the time that it is disconnected will unfortunately not be pulled into Analyze.

  4. We are not able to backfill follower counts for Instagram. Instead, we provide an approximation of the previous growth on the account based on the current growth. To do this, we will look at the follower count from the past day, as well as the current follower count, and generate a graph of what growth could have looked like in the past (found in the Overview tab). If you experienced larger fluctuations in follower counts (e.g. losing or gaining 100 followers) in a single day, the graph would not represent that and would instead reflect a smoother estimate based on the data we’re able to see when the account was connected.

  5. Audience information for Instagram will take up to 24 hours to populate.
  6. We are not able to pull in data on IGTV posts.

  7. Data on individual Instagram posts will not be updated indefinitely. After a full calendar month has passed, we will no longer update the metrics for that post.

  8. The Instagram stories displayed will only be those that are original content. Analyze will not display any information on reshared stories.

Twitter

  1. 28 days of Twitter data is backfilled when you connect your account to Analyze. This means that your data won’t match what you see in Twitter Analytics when you’re looking at a time period 28 days prior to the account being connected. Let’s say your Twitter account was connected on October 1, 2019. In January 2020, if you looked at your data in Analyze for the whole year of 2019, it wouldn’t match what you see in Twitter Analytics for 2019 because we are only capturing four months of data (the 28 days prior to connecting the account, which is the data we’ve backfilled, and the three months after connecting the account).

    📝 Please note, if you are also subscribed to the Premium or Business plan for Buffer Publish, and your Twitter account is connected there, this backfilling won’t be applicable to you. In that case, since we’ve already gathered metrics for your Twitter account for Publish analytics, those will be copied over to Analyze, and will go as far back as we’ve been gathering data there.

    If you were previously using the Premium or Business plan for Buffer Publish (either through a trial or a paid subscription), assuming you didn’t fully delete your account, we would still have that data saved for your Twitter account. In that case, when connecting the account to Analyze, we would copy over that data and backfill the past 28 days. This could result in you seeing gaps in data over certain periods of time in the past.

  2. We are not able to backfill follower counts for Twitter. Follower counts will start populating when you connect your account to Analyze.

  3. Data on individual tweets will not be updated indefinitely. After 30 days have passed, we will no longer update the metrics for that post.

  4. The number of tweets you’ve shared, which is shown in the Posts tab and the Overview tab, will not include replies or retweets. Twitter Analytics do include replies and retweets, so these numbers may differ.

  5. The engagement and engagement rate stats you see in Analyze will likely be a bit lower than what you see in Twitter Analytics. This is because the Twitter API version that we are using does not offer an aggregated engagements number for each tweet. Therefore, we are calculating the engagements ourselves, using a sum of likes, replies, retweets, quotes, link clicks, and profile clicks. Other engagement types, such as app opens, app install attempts, and hashtag clicks, are not included. The engagement stats in Twitter Analytics on the other hand, will include all types of engagement.

Other articles that might be helpful

Was this article helpful?
4 out of 12 found this helpful