Evaluating Promotional Efforts for Driving Traffic to an Extension Outreach Website

Introduction

Online communication is integral to how people today share experiences and information. Eighty-eight percent of American adults are now online (Pew Research Center, 2016). Moreover, most online adults are using some sort of social media platform; nearly 80% use Facebook, and a third use tools such as Instagram, Pinterest, LinkedIn, or Twitter (Greenwood, Perrin, & Duggan, 2016).

Extension professionals have realized the importance of online communication and are using these tools to communicate with their communities. Several articles in the Journal of Extension have described social media tools (Kinsey, 2010), discussed strategies for increasing engagement (Israel et al., 2015; Mains, Jenkins-Howard, & Stephenson, 2013), and offered suggestions for collecting campaign metrics (Christensen, Hill, & Horrocks, 2015; Gharis & Hightower 2017). However, less is known about how different promotional strategies affect engagement with online content.

To explore this concept, we created a website (StopTheSpiny.com) that features an unconventional outreach video that blends the arts, natural sciences, and social sciences with information on how to prevent the spread of the invasive spiny water flea (Bythotrephes longimanus). This was determined to be a timely topic by natural resources managers because spiny water fleas are difficult to see, are in the early stages of the invasion process, and have been shown to have large negative economic and ecological impacts (Walsh, Carpenter, & Vander Zanden, 2016). By combining promotional efforts with tracking of website traffic, we were able to determine how effective certain promotional efforts were in driving visitors to our web-based content. An understanding of the approach we used can help Extension professionals move beyond reporting reach and impressions to reporting desirable behaviors. Although our topic area was invasive species, we believe our results are applicable across Extension program areas.

Methods

The website and the video were created in early 2016. We intentionally designed the video to differ from traditional didactic video content by including special effects, references to pop culture, and elements of viral online content. Our hope was that this approach would increase the number of website visits to more than we might have received with a more traditional outreach campaign.

After completing production of the outreach materials, we developed our communications plan, which included the elements identified in Table 1. In our outreach techniques, we incorporated standard tools used by Extension agents—press releases, email lists—as well as newer tools, such as boosted Facebook posts and Google AdWords.

Coordinated online promotion efforts consisted of existing activities Extension agents use to promote their materials, including sharing press releases with partner organizations via social media (e.g., Twitter, Facebook) and electronic mailing lists. We also worked with on-campus communications departments to promote our message. Through this approach, we reached internal audiences and people we considered opinion leaders in the fields of aquatic invasive species and science communication who would distribute our messages further. We spent 5 hr coordinating with partners on these efforts.

Our earned media efforts consisted of speaking to any journalist or media outlet interested in disseminating information about our video and invasive species in general. We spent about 5 hr preparing for and completing the interviews.

A boosted Facebook post is shown only to users who fit specific demographic and interest criteria. They are not likely to be actively seeking Extension content (for example) but may be interested in it. For our three boosted posts, we paid to have posts with text and a short video shown to males aged 18–65 who indicated interest in outdoors, fishing, angling, recreational fishing, boating, or lakes. We limited the sample to only men due to previous work suggesting that a vast majority of boat owners in Wisconsin are male (Witzling, Shaw, & Seiler, 2016). For two of the posts, we limited the geographic scope to Facebook users in Wisconsin. For one post, we targeted residents of a specific county that has multiple lakes invaded by spiny water flea. The time to set up and manage this method was less than 30 min.

Google AdWords displays a text advertisement, including a link to a website, as a result of a Google search. Advertisers are able to place advertisements on keyword/key phrase searches relevant to their target audience and pay only for the number of clicks an advertisement receives. The audience consists of active information seekers and, depending on the keywords, can include people searching for the advertiser's topic or related topics. We used more than 40 keyword that related to boating, fishing, or spiny water flea waters. The time to set up and manage this method was less than 30 min.

We used Google Analytics to track the number of website visits attributable to each of the promotional methods. Google Analytics is a service that tracks website visits, how users get to a website, and data associated with the visits. For our analysis, we used only the raw number of visits during a particular time period. The website visits reported in this article were from June 2016 through October 2016 (for coordinated online promotion, earned media, and boosted Facebook posts) and May 2017 through September 2017 (for Google AdWords). We never had more than one promotional method in use at a time during each of the study periods, so we attributed a website visit to whichever promotional method was occurring at the time. Although demographic data can be collected using Google Analytics, we did not collect such data as part of the project.

We calculated cost per site view differently for each method. For coordinated online promotion and earned media, the primary cost is time spent on the effort. The estimate would be dependent on the wage of the person doing the task and the strength of that person's network. However, so that we could make comparisons across the promotional methods, we estimated cost based on 5 hr of work for each method and a $25/hr wage, including benefits. For boosted Facebook posts, we divided the total dollars spent on the promotion by the number of attributable website visitors during the promotional period as determined by Google Analytics. For Google AdWords, we received cost-per-site-view data directly from Google AdWords.

It is possible that people were exposed to multiple promotional methods over time given the methods' sequential nature. However, we believe that each method targeted different audiences and that those exposed to multiple promotional methods were in the minority.

For Facebook posts, we did not track reach, impressions, and other engagement metrics as our goal was to determine how to drive people to a website. Future work could address the relationship between promotional methods and metrics related to reach and engagement (e.g., reactions, shares).

Findings and Implications

All promotional methods drove visitors to StopTheSpiny.com. However, differences existed related to effort, cost of the promotion, and number of visitors driven to the website. Table 2 shows the approximate number of website visitors and cost per site view for each method of promotion, and Figure 1 shows the visits over time. The sections following those graphics describe our lessons learned from using each method.

Figure 1.
StopTheSpiny.com Site Visits Associated with Different Promotion Efforts Over Time

Coordinated Online Promotion

The biggest benefits of the coordinated online promotion approach were that it was free, outside of staff time, and it drew an initial surge of website visitors. If we include staff time, this was still the least expensive promotional option. The drawback was that this effort was one of the most time intensive—as noted, we spent 5 hr coordinating with partners. On the other hand, an upside is that such coordination can lead to additional benefits through stronger connections with partners.

Earned Media

Because we approached an invasive species issue in an unconventional way, we were able to earn stories in mass media outlets. We earned print and online articles by a renowned outdoor writer in a large regional newspaper, a radio interview and an online article with Wisconsin Public Radio, and an online article with New York Public Radio. These stories allowed us to discuss both the impacts of the spiny water flea and the use of unconventional outreach methods.

Earned media appeared to drive a similar number of visitors to StopTheSpiny.com as the coordinated online messaging effort (see Table 2), but likely reached a different audience. We believe that the earned media placements, such as coverage by a respected outdoor writer, reached more of our target audience, whereas our coordinated online effort likely reached internal audiences (e.g., university colleagues, invasive species professionals). In addition to website visits, these media placements resulted in hundreds of thousands of exposures to our message. A drawback to general use of this approach is that obtaining earned media placements for an outreach campaign typically would occur only if the topic were especially timely or the campaign approach unconventional.

Boosted Facebook Posts

The boosted post feature displayed our post to tens of thousands of people for $121.29 (less than $0.01 per exposure). Far fewer people actually visited StopTheSpiny.com by clicking the advertisement; cost per website visit ranged from $0.76 to $1.25 across the three boosted posts. However, cost per site visit was likely lower than what was reported. Google Analytics data showed that many people directly navigated to the site as opposed to being directed there via Facebook while boosted posts were running, but those visits were hard to link with the advertising effort and thus were not considered in calculations of cost per site visit. Both direct views from our advertising and views resulting from users sharing our post factored into our calculation. We cannot separate the two due to the way in which Facebook reports reach results.

Getting people to take an action outside the Facebook ecosystem (i.e., click through to StopTheSpiny.com) was difficult. Because of this, we believe that boosted Facebook posts are best used for displaying messaging to large numbers of people or enticing people to take action within Facebook (e.g., commit to an event, "like" a page). We believe that including the relevant content within a Facebook post itself, as opposed to having users click to leave Facebook, would boost efficacy in message engagement. It is worth noting that we used a few different approaches (e.g., different messages, geographic targeting) in our advertising and noticed some variability in the cost per click. We believe this variability suggests that there are opportunities to optimize messaging to increase click-through rates and reduce the cost of Facebook paid advertising if that is a primary goal.

Google AdWords

Use of Google AdWords was an efficient and relatively inexpensive method of generating visits to StopTheSpiny.com. The tool took a minimal amount of time to set up and administer (<30 min), provided a large amount of data on the efficacy of the advertising, and required essentially no additional effort after the initial setup. The average cost was $0.45 per website click, with topic-specific search terms being less expensive due to lower cost for better advertisement placement (i.e., higher positioning in search results) and the likelihood that the advertisement was more relevant to those who clicked through. However, the larger number of people searching using more general search terms resulted in more clicks (Table 3).

Because AdWords advertisements are displayed to thousands of people, even those who do not click the advertisement, there is the possibility of additional benefit through brand impressions. Brief and thoughtful wording can help users connect with the message, even if they do not click the advertisement. However, these impressions are likely less valuable here than on Facebook given the inability to add visual media to the AdWords text advertisement.

Lastly, we believe there is an opportunity to advertise inexpensively on search terms that do not currently have paid advertising on them. By using this tactic, for a relatively low cost, one can make relevant content display on the first page of a keyword search. In our example, terms highly specific to our topic (spiny water flea, Lake Mendota [a lake that contains spiny water flea]) were effective choices because few people were advertising using these terms and people searching for these terms were actively looking for information on topics that could be affected by invasive species. This approach could be especially useful if the website being promoting is not on the first page of search engine results.

Conclusions

Many Extension practitioners have communicated the importance of using online communication tools. Although all forms of promotion can be useful, selecting the best technique can increase the efficiency of efforts to make partners aware of content, reach different audiences, generate a large number of exposures, or drive traffic to a specific website. Approaches involving coordinated online promotion and earned media can be most cost-effective during the period of initial interest that occurs when a project is new. However, as the initial interest wanes, efforts may need to shift to other methods of promotion, such as boosted Facebook posts or Google AdWords online advertising, to keep people interacting with content. The results reported here should help those promoting an Extension website obtain visitors in a cost-effective manner.

References

Christensen, A., Hill, P., & Horrocks, S. (2015). The social media marketing map (Part 1): A tool to empower the digital leaders of Extension. Journal of Extension, 53(4), Article 4TOT3. Available at: https://www.joe.org/joe/2015august/tt3.php

Gharis, L., & Hightower, M. (2017). A practical method for collecting social media campaign metrics. Journal of Extension, 55(2), Article 2IAW6. Available at: https://www.joe.org/joe/2017april/iw6.php

Greenwood, S., Perrin, A., & Duggan, M. (2016). Social media update 2016. Retrieved from http://www.pewinternet.org/2016/11/11/social-media-update-2016

Israel, G., Borger, R., Greer, K., Kelly, S., Byrum, K., Pelham, J., & Momol, T. (2015). Using Facebook advertising to connect with Extension audiences. Journal of Extension, 53(4), Article 4FEA10. Available at: http://www.joe.org/joe/2015august/a10.php

Kinsey, J. (2010). Five social media tools for the Extension toolbox. Journal of Extension, 48(5), Article 5TOT7. Available at: http://www.joe.org/joe/2010october/tt7.php

Mains, M., Jenkins-Howard, B., & Stephenson, L. (2013). Effective use of Facebook for Extension professionals. Journal of Extension, 51(5), Article 5TOT6. Available at: http://www.joe.org/joe/2013october/tt6.php

Pew Research Center. (2016). Internet/broadband fact sheet. Retrieved from http://www.pewinternet.org/fact-sheet/internet-broadband/

Walsh, J., Carpenter, S., & Vander Zanden, J. (2016). Invasive species triggers a massive loss of ecosystem services through a trophic cascade. Proceedings of the National Academy of Sciences, 11(15). Retrieved from http://www.pnas.org/content/113/15/4081.abstract

Witzling, L., Shaw, B., & Seiler, D. (2016). Segmenting boaters based on level of transience: Outreach and policy implications for the prevention of aquatic invasive species. Biological Invasions, 18(12), 3635–3646. doi:10.1007/s10530-016-1254-7