Analysis of Facebook Content Marketing for a Non-Profit

man overwhelmed by social media

This article is part of a larger and ongoing collection of articles about content marketing for a non-profit organization. I invite you to read the “parent” article, Content Marketing Strategy for a Non-Profit: A Case Study.

"You simply must be on social media!" This has been said so often, by so many people, with such varying levels of expertise, that it has become something of a cliché. Snarkliy, perhaps, I have come to refer to such statements as social media cheerleading: "Rah, rah, sis boom bah! Social media to the rescue!"

Of course, that statement contains truth. Brands, including non-profits, do need to be present and active on social media. But a social media strategy is a little easier said than done. And, one of more difficult aspects of social media marketing is evaluating the effectiveness of your campaign.

This article analyzes, in some detail, a Facebook campaign carried out for a non-profit career center. It is my hope that this analysis will assist other non-profit organizations as they consider their own Facebook marketing campaigns.

Background and Strategy

The organization was a federally-funded career center that, as the economy slowly recovered, was experiencing a decline in job seekers pursuing their services. Indeed, through 2013 and 2014, the local unemployment rate fell from 8.1% to 4.9% – this represented a 40% drop in potential "customers".

The Facebook campaign was part of a larger social media campaign, which was part of a larger content marketing campaign to drive awareness and engagement with job seekers.

The social media campaign was undertaken with the knowledge that a limited amount of time each week could be invested in social media activity, and with the related understanding that this would limit the results obtained. Also, paid promotion of posts was out of the question; all reach would need to be achieved organically through the quality of content posted.

Content for the Facebook page (as well as other social platforms) would be a mix of content, some curated from other sources, some created in-house. Overall, content for Facebook posts fell into these categories:

  • Articles, curated from other sources
  • Small infographics created in-house (usually based on survey data from major 3rd party sources, e.g., CareerBuilder)
  • Links to original content on the career center’s website
    • As a special subset of this, links to the popular Job Search Dashboard
  • Job fairs held in the career center
  • The center’s biweekly Job Club
  • Jobs posted on the career center website
  • General PR / success stories

Based on website traffic patterns, I predicted that optimum days to post would be Tuesday, Wednesday and Thursday, since traffic to the website was markedly higher on those days. This graph represents typical weekly traffic to the website:

Graph of weekly visits to website

Anecdotal evidence from recruiters also confirmed that job seekers were most active Tuesday through Thursday. Additionally, experiments with posting content on weekends further confirmed that job seekers are considerably less engaged on weekends. Therefore, frequency of posts was 3 to 5 days every week, usually Tuesday through Thursday, usually one post per day

Results and Analysis

The analysis was based on Page data, as well as data for individual posts, downloaded from Facebook for the two year period January 01, 2013 through December 31, 2014.

Page Data and Analysis

Page likes grew continually, and at a very steady pace, approximately tripling over the two year period:

Graph of Facebook page metrics

 
Reach and engagement on the page also grew steadily, showing a marked increase in activity for the period Jan 1, 2014 through July 1, 2014. After this there was a noticeable drop off. It is assumed that these fluctuations were due to changes to Facebook’s algorithms, as there was no significant change in content publication corresponding to these periods.

Graph of Facebook page metrics

The significant peak in Reach obtained in February 2014 was a post related to workshops held in the career center for people laid off from a very large and high-profile company that closed in the area.

Post Data and Analysis

Average post reach showed an overall trend upward. The very high value obtained in Q1 2014 was due to the layoff-related post mentioned above:

Graph of overall Facebook post reach, by quarter

 
Average Click Rate was high in the first half of 2013, then dropped. This decline is attributed to the fact the significant drop in local unemployment, which represents a decreasing number of people actively seeking job search information and assistance. However, despite that decline, there was an overall increase in engagement with the content; overall Engagement Rate increased to 7.75% in 2014, from 6% in 2013. This is all illustrated in this graph:

Graph of overall Facebook post engagement, by quarter

Analysis by Post Type

In an attempt to dig deeper into patterns of user engagement, data for different types of posts was examined.

Curated Article posts – Engagement for articles curated from third party sources remained steady over the two year period, with two notable patterns: 1) A decline in Clicks, again attributed to the declining in unemployment rate, and 2) a notable spike in Reach toward the end of 2014.

Reach and engagement for news article posts

 
Infographic posts – Posts of custom infographics achieved the lowest Reach and Engagement numbers, and also showed overall declines across all metrics:

Reach and engagement for infographic posts

 
Jobs posts – Posts about jobs listed on the website showed a similar pattern to the Dashboard, i.e., steady Engagement, a drop in Clicks, but an increase in Reach:

Reach and engagement for jobs posts

 
Job Club posts – posts about the bimonthly job clubs again showed the patterns of steady Engagement, a decline in Clicks, but an overall increase in Reach:

Reach and engagement for job club posts

 
Job Search Dashboard posts – These posts showed a similar pattern, stead engagement with a decline in Clicks but an increase in Reach through 2014:

Reach and engagement for dashboard posts

 
Website Content posts – There is less data for these posts, but they showed a similar pattern, with the exception of a spike in Reach for a particular post in June 2014., which was a post about professional certifications in I.T.:

Reach and engagement for  posts of website content

The above data also confirms the growing importance of mobile-optimized content: the two highest peaks were for mobile-optimized content.

 
Posts with Photos, versus No Photo – The graphs below show differences in Reach, Clicks and Engagement for posts with and without photos. Data was analyzed for all post types, and also specifically for Job Search Dashboard posts. Dashboard posts were a focus because these were posted often, sometimes with a photo, and sometimes without.

The first graph shows, in line with the conventional wisdom, that posts with photos achieved significantly higher Reach:

Effect of photos on overall post reach

 
The second graph shows Clicks and Engagement for all posts, as well as for Dashboard posts. The results were very interesting. In general, across all posts, Click rates increased when a photo was used. This is in line with conventional wisdom regarding use of photos. However, for Dashboard posts, photos appeared to actually reduce Clicks. Across the board, for any and all posts, engagement was not affected by the used of a photo. This is all illustrated in this graph:

Effect of photos on overall post engagement

The anomalous Click rate for Dashboard posts may be explained by the fact that the Dashboard implements an Open Graph image tag, so that posted links to the Dashboard (without a photo) still present an eye-catching illustrative "icon".

The most disappointing outcome, was the decline in Facebook referrals to the organization’s website:

Graph of Facebook page metrics

Given the considerable drop in unemployment (cited above), this was not surprising, but it did suggest that the organization would benefit from more frequent posting of links to various landing pages on their website.

Conclusions and Next Steps

Given the very limited amount of time that could be invested in the Facebook campaign, it has been deemed successful for the following reasons:

  • The Page is gaining new Likes at a steady and ongoing pace.
  • While Clicks and Engagement declined in 2014 (symptomatic of the relatively low unemployment rate cited above), overall post Reach increased.
  • Facebook continues to be a top source of social referral traffic to the organization’s website.

Next Steps to be Taken:

  • Continue the mix of content currently being offered, however…
  • Increase the percentage of posts offering information other than direct job search advice (since, again, the number of people seeking such information has fallen so dramatically). Such posts would function much more in the realm of PR and awareness-raising, i.e., events, success stories, etc.
  • Post links to website content more frequently.
  • Be certain that posts always include a photo, or at least an Open Graph image.
  • Continue migrating website content to a mobile-friendly format.