Social Media Analytics, Part 2

By: Dr. Kathy Kelley

In last month’s blog, Social Media Analytics, Part 1 (http://bit.ly/1Q4Htsz), I provided information about data available on Twitter Analytics (https://analytics.twitter.com) based on Denise Gardner’s Twitter profile/account (@DeniseMGardner).  For example: number of followers, follower response (number of impressions, mentions, etc.), and days of the week/times of the day that she might want to schedule her tweets.

In addition, I included a list of reasons why Twitter users tend to unfollow others and why you might want to post a tweet more than once. In this week’s blog you will find information about how to find key Facebook metrics using Denise’s Penn State Extension Enology Facebook Page as the example.

What Information does Facebook Insights Provide?

If you are not familiar with Facebook Page “Insights,” like Twitter Analytics, they provide information that can help you determine who “likes” your page, when you might want to post to your Facebook Page, and what posts Facebook users interacted with the most.

You can view these data “after at least 30 people like your Page” (http://bit.ly/21vnFoS) by clicking on the Insights tab at the top of the page.

Apr 2016_KathyB_Image 01

All of the information discussed in today’s blog can be found by selecting options such as “Overview,” “Likes,” and other categories found on the Insights tab (see image below).

Apr 2016_KathyB_Image 02

There is a fair amount of data provided via Insights and on first view it can be overwhelming.  But, the information provided below can help you understand some of the definitions (e.g. consumption) and differences between terms (e.g. engaged user vs. consumer) used to describe the categories of data.

Data you can access on Insights “Overview”

Like it sounds, Overview is the first section that you should look at to learn about Facebook user response to your posts.

1) Number of page views

Page views, as you might expect, are the number of of times people viewed the Penn State Extension Enology Facebook Page.  Denise can change the graph (below) to show data specific to “today,” “yesterday,” the “last 7 days,” and the “last 28 days.”  In addition, she can export Page, Post, or Video data as an Excel or .csv file for any time period (“lifetime” of the Facebook Page to “today”) that she specifies.

Along with numbers, a percent change (positive or negative) shows how the current number of page views compares to an earlier point in time.  For example, in the image below, you can see the number of page views for the period of March 25 to April 11, 2016 (138).

Picture1

Notice that there is a green upward arrow and “27%,” which is the increase in page views compared to the previous 28-day period.  If I were to change the range to “the last 7 days” you would see the total/percent for the period of April 5 through the 11 as compared to the previous seven days.

2) Page Reach

Page Reach “is the number of people who saw any of your post content during a give period of time” (http://bit.ly/1pUi5we).

For the period of March 15 to April 11, the total number of Facebook users who “saw” content on Denise’s Facebook Page was 4,496; however, this number was 35% lower than the page reach for the previous 28 days.

3) Page Likes

This metric refers to the number of Facebook users who “liked” Denise’s Facebook Page during the period specified.  As with Page Views and Page Reach, the number of Page Likes gained between March 15 and April 11 is expressed as a number and a percent increase/decrease.

4) Reach and Engagement for Individual Posts

Just below these graphs you will find a table that highlights the five most recent posts.

Picture2

In the table, you will find:

  • The publication date for each post;
  • post title;
  • type of post (link, photo, video, status update);
  • targeting, which means “who” the post was shared with (e.g. public, which would be anyone on or off Facebook; your friends on Facebook; customized targeting);
  • total reach (6,475 people)
  • organic (3,025) and paid (3,450) reach
    • by “hovering” your curser over the data you will see how many were organic (those who saw the post “through unpaid distribution”) and paid (those who saw the post as a result of an ad/post was “boost,” http://bit.ly/1SUbEEJ); and
    • engagement
      • post clicks (166) and the combined number of all reactions/comments/shares (114) for the post.

Denise’s post on March 11, 2106 “reached” nearly 6,500 people.  Since her posts are shared with the public (notice the globe symbol), anyone “on or off Facebook” can view them. If she wanted to “customize” who sees the post (e.g. audience location, language, gender, age range) she could do so by changing the page settings.

Data you can access on Insights “Posts”

1) Additional Reach Data

Step 1: Click on the arrow on the top right of the table to be directed to Insights “Posts” to see additional/different breakdowns of reach and engagement.

Apr 2016_KathyB_Image 04

Step 2: On Insights Posts, click on the drop-down arrow to the right of “Reach: Organic/Paid” (highlighted by the red circle in the following image). There you will find:

  • the number of Penn State Extension Enology Facebook Page “fans” (371) and number of non-fans (6,104) who saw the post and
  • organic impressions (posts “displayed in a user’s News Feed, ticker, or on [the Extension Enology Facebook] Page”) and paid impressions (“number of times your paid content was displayed,” http://bit.ly/20ZZEYn).


Apr 2016_KathyB_Image 05

Note that “people might see multiple impressions of the same post…in [their] News Feed once…and then a second time if their friend shares it” (http://bit.ly/1wrk7HQ), so impressions are not based on “unique” views.  Hence, this metric may not be as useful as the other reach/engagement data.

2) Additional Engagement Data

Clicking on the drop-down arrow to the right of “Post Clicks, Reactions, Comments & Shares” (red circle in the image below) we can see a few options as to how the data can be presented:

  • engagement, which includes number of post clicks (166) and the combined number of reactions, comments, & shares (114);
  • number of post hides, hides of all posts, spam reports, page unlikes; and
  • engagement rate (3%), which is calculated based on number of people who the post reached who then liked, commented, shared, and clicked on the post.

Apr 2016_KathyB_Image 06

You may notice that posts with greater reach do not necessarily have the highest engagement rate.  Engagement through videos tends to be higher than photos, post links, and text only posts (http://mklnd.com/1FVSmdK).  This same article provides a response to the question: “What is a good engagement rate on Facebook?”  Of course this varies based on industry, followers, and other factors, but you could compare each of your posts’ engagement rate against the following:

  • “Above 1% engagement rate is good,”
  • 0.5% to 0.99% “is average,” and
  • below 0.5% “likely means that you need to realign your messages to that of your audience’s expectations…more compelling and engaging contributions.”

Boosted Posts

Several articles have indicated that organic reach “is steadily declining” and that ads and boosted posts are becoming even more important to businesses trying to attract Facebook users and build their clientele base (http://bit.ly/1fJnS5R).

Boosted posts will appear in a Facebook user’s News Feed, and according to one blog (http://bit.ly/1KbmLFn), “paid reach on posts drives additional organic and viral reach/engagement” and “residual organic effect” occurs “for a few days” after running an ad.  So, after the boost/ad period expires you may see reach gradually taper off rather than drop suddenly.

Options for who can see your boosted posts are:

  • People who like your Facebook Page,
  • people who like your Facebook Page and their friends, or
  • an audience you target based on their location (e.g. country, state, city, distance from a specific address), gender, age range, and four to 10 interests (e.g. wine, wine tasting, tasting, live music, live music festivals; http://bit.ly/2104Onk).

You then indicate the amount you want to spend and the duration you want the campaign to run.  Facebook will then provide an estimate of the number of people that you will reach based on the audience you are targeting and the budget.

Another option is to use the Ads Manager feature and “promote a post,” which provides even more targeting, pricing, and bidding options (http://bit.ly/1fJnS5R).

Using the Data Export Feature

To view data for several posts, use the Data Export Feature to download the information to either an Excel file (.xls) or a Comma-separated values (.csv) file.  Begin the process on navigating to Insights “Overview” and then click on Export and select the data type (Page, Post, or Video data) and data range.

Apr 2016_KathyB_Image 07

I exported the “Post Data” for January 1 to 31 March, 2016 and on the “Key Metrics” tab I sorted the posts by Type (status, photo, link, or video), Lifetime Posts Total Reach (the total number of unique users that the post was “served” to since being published), Lifetime Post Organic and Paid Reach, and many other metrics.

Post Reach

The following image shows a sample of the data I sorted by “Lifetime Post Total Reach.”  Of the posts published on the Penn State Extension Enology Facebook Page between January 1 and March 31, 2016, the March 11th post had the highest total reach: 6,475, and (as you also saw above) 3,025 of these people were reached organically and 3,450 were reached because the post was boosted.

Of the six posts shown below, this was the only one that was boosted, so that is why all other Lifetime Post Paid Reach cells have a “0.”

Apr 2016_KathyB_Image 08

Consumers and Engaged Users

Facebook provides data that describes Consumers and Engaged Users (http://bit.ly/1ggqTfj).

The number of Consumers is calculated based on Consumption:

  • Link clicks,
  • photo views,
  • video plays,
  • post comments, likes, and shares.

Engaged Users include the number of these consumers and those who clicked on:

  • Name of the user who commented on the post,
  • number of likes a comment receives,
  • responses to comments, and similar.

If this is unclear, refer to the image, below.  This visual shows what types of actions count as “consumption” (clicks/comments/sharing/likes in the “blue” rectangle) and how the number of engaged users is calculated (consumption actions in the “blue” rectangle and actions that take place in the “green” rectangle; http://bit.ly/1iUuqgZ).

Apr 2016_KathyB_Image 09

Since the number of Lifetime Engaged Users includes the number of Lifetime Post Consumers and other clicks, the number of Lifetime Engaged Users will always be equal to larger than the number of Lifetime Post Consumers (http://bit.ly/1ggqTfj).

The following image shows data for the top six posts and the location of these data in the Excel spreadsheet: Lifetime Engaged Users can be found in Column “L” and Lifetime Post Consumers can be found in Column “M.”

Apr 2016_KathyB_Image 10

Reach and Engagement for Individual Posts

If you access Insights “Posts” and then click one of the title, you will see People Reached and Post Clicks.

For a post written by Michela Centinari, 813 people were reached and there were 55 Post Clicks.  Of these Post Clicks, 36 were Link Clicks and 19 were Other Clicks (clicks on other user’s names, their comments, and on content other than a link/photos/videos).

Apr 2016_KathyB_Image 11

Likes (24), Comments (4) & Shares (4) can be divided further into those that occurred “On [the] Post” and “On Shares.”

Apr 2016_KathyB_Image 12

Looking at the “blue” rectangles above, 12 users clicked on the post’s “like button,” two provided comments (one of which was Denise’s reply to the initial comment), and three shared the post with their friends and fans.

As a result of three fans who shared the post with their friends/fans (highlighted by the “red” rectangle): 12 of their friends/fans clicked on the post’s like button; two of their friends/fans commented on the shared post; and 3) one of their friends/fans further shared the posts with their friends/fans (data highlighted by the red rectangle).

How Many People Liked your Facebook Page on a Given Day?         

Obtaining this data is pretty straight forward.  Insights “Likes” provides a visual that, by “hovering” your curser over it – a pop-up appears with the number of Total Page Likes for that day.  Several options exist pertaining to the range of data that can be shown: a week, month, quarter, or a specific date range.

Apr 2016_KathyB_Image 13

Interested in learning the total number of likes your page had on March 11, 2106, or perhaps the number of additional likes you earned on that day?  The Data Export Feature can be used to download Page Data.  I’ve provided a condensed view of the Excel file with the data for the March 11 post and the column in the file where you will find the data.

Apr 2016_KathyB_Image 14

Examples of what you will find for the page include:

  • Lifetime Total Likes (536 for the March 11 post; column B),
  • Daily New Likes (4; column C),
  • Daily Unlikes (0; column D),
  • Daily Page Engaged Users (83; column E),
  • Daily Paid Reach (384; column N)
  • Daily Organic Reach of Page Posts (1,344; column AG), and
  • Daily Paid Reach of Page Posts (384; column AJ).

The percent of your fans who are male and female and in each of the reported age ranges (e.g. 18 to 24, 25 to 34 year), along with gender/age range breakdown of “People Reached” and “People Engage” can be found on Insights “People.”

Apr 2016_KathyB_Image 15

Along with this information, you can learn how many of your fans “were on Facebook” during the previous seven days and what time of day seemed to be the “most popular.”

Apr 2016_KathyB_Image 16

The number of Penn State Extension Enology Facebook Page fans didn’t change much from from day-to-day for the seven-day period shown (range of 506 to 514), and at least 200 fans checked Facebook between the hours of 8 a.m. and 10 p.m., with 270 on Facebook at 9 p.m.

As with Twitter, you should experiment with the timing of your posts.  Look at the resulting response and engagement to determine when it is optimal to post.  According to one author, consider posting at “off-peak” hours “when fewer people are sharing content on Facebook” in hopes that more fans will notice your post (http://bit.ly/1PZ4Cmv).

Where did your Facebook Page Traffic Originate from?

Perhaps you are using several outlets (your website, a link in an email, other social media) to reach your customers and drive them to your Facebook Page.  Insights “Page Views” provides a visual that shows how many visitors originated from your blog, Google, those who were already on Facebook, etc.

The following graph shows the number of views on April 7, 2016 that originated from four sources.   Seven people were already on Facebook, one was redirected from the Penn State Extension Enology WordPress blog, and another from the Penn State Extension website.  While none of the views on the 7th of April originated from Google, one did originate from Google on April 6th.

Apr 2016_KathyB_Image 17

This information can help you determine where to focus marketing and promotion efforts to increase traffic on your Facebook Page.  If Denise was unaware that traffic was generated from the Penn State Extension website she might consider contacting the administrator and ask if she could provide content that could be posted on the site with a link to her Facebook page.

The third blog in the Social Media Analytics series will focus on WordPress and Blogger analytics.

A special thank you (again) to Denise Gardner for allowing me to show all her analytics in this post!

Advertisements

Tags: , , , , , , , , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: