SAS Social Media Analytics: Monitoring a Brand with a Brand
On November 7, 2010, KDPaine Partners, a leading public relations and social media measurement firm, and SAS, the leader in business analytics software and services and the largest independent vendor in the business intelligence market, won the prestigious New Communications Award of Excellence for their breakthrough work in social media research.
According to KDPaine, "SAS and [KDPaine Partners] won the award in the Social Data Measurement/Measurement Innovation Division for their work transitioning traditional PR measurement into an integrated, consistent and comprehensive communications measurement program….The research program gathered information from thousands of traditional and social media outlets, including YouTube, Facebook, social bookmarking sites, as well as internal and external blogs. Each item was analyzed to determine the nature of the conversation, content of the posting, engagement level, and how the author positioned the company."
SAS is the leader in business analytics software and services and the largest independent vendor in the business intelligence market. Through innovative solutions, SAS helps customers at more than 50,000 sites improve performance and deliver value by making better decisions faster. Since 1976 SAS has been giving customers around the world The Power to Know®.
As a leading provider of business analytics solutions to organizations worldwide, SAS has long understood and valued the role that media, traditional as well as social, plays in the sales cycle. For more than a decade SAS has monitored and measured its competitive position in the media. With the rapid rise in influence of social media in the last two years, it became clear that the measurement and metrics program needed to be expanded to include social conversations. It was also clear that social media offered an entirely new opportunity to measure the success or failure of communications efforts through correlations with web activity. The challenge was how to accurately reflect and report on SAS' results in order to provide data on which to make better decisions. And specifically, would the metrics be able to be used to better understand what worked and didn't work around the launch of SAS' own SAS® Social Media Analytics product in April at their annual SAS Global Forum international user conference.
Issues To Be Addressed
The fundamental goal for SAS was to create a reliable, accurate and useful set of metrics that would help marketing, PR and communications professionals make better decisions. The metrics had to meet a rigid set of requirements including;
1. Accuracy of data-without management's trust in the data, the reports wouldn't be valuable. This was complicated by SAS' very name, given social media's predilection for abbreviation. One SAS employee has collected more than 400 unrelated "SAS" acronyms via Google alerts alone.
2. Timeliness of data-social media has increased the speed of mass communication, and thus demands reduced response times and quicker decision making. Not only did the data have to be available 24/7/365, but it also needed to be processed-read and coded-in record time if it was going to be effectively used to shape strategy.
3. Context and clarity-SAS didn't just need numbers; they needed to understand why the numbers came to be what they were. They needed to understand the competitive environment in which they were operating, and they needed to quickly understand what the data meant.
Define the Target Audience
The SAS PR team's traditional target audience has been IT decision makers-CEOs, CIOs, CFOs and senior executives in marketing and operations-in Fortune 1000 companies. But with its new SAS® Social Media Analytics-a SAS Solutions OnDemand offering-the audience instantly became much broader. Any organization that needed to better understand its social media program-including employees at various levels within PR and advertising agencies, consumer package goods, services companies, retailers, and financial services firms across the board-essentially anyone working with customers was now part of their target audience.
Goals and Objectives
With the above concerns in mind, SAS, in conjunction with KDPaine Partners, designed a research program. The following major goals were decided upon:
1. Provide ongoing, accurate and timely data to help SAS make better decisions regarding the direction of external communication activities.
2. Correlate the data with online activity to determine levels of engagement and outcomes from specific programs
3. Advise the social media team on what it should or should not be doing in social media: What changes should it make to its present programs? What new programs should it add?
4. Set benchmarks against which future programs could be judged.
Step 1: Data Collection
The first step in helping SAS achieve its goal of understanding social media was to standardize collection techniques for SAS and its competitors. The biggest challenge was simply getting relevant data. So much of the social media conversation happens in 140 characters; SAS can easily become an abbreviation for everything from Surfers against Sewage to Second Avenue Subway. While automation and exclusion tables helped, they posed a challenge as well: set your exclusionary tables too strictly and you miss relevant information. Set them too broadly and you drown in data. It was decided to observe and explore a predefined range of social media channels for SAS and its competitors, and then define a list of the top 100 channels or sites that would be read in depth. Typical patterns of traffic and usage could then be determined and used as a starting point for understanding where the institutions stood with respect to this new media.
Step 2: Standardizing Qualitative Data
As with any content analysis project, [KDPaine SAS] also needed to establish standardized definitions of tone, positioning, and visibility. By establishing these metrics SAS would be able to see just how it and its peer institutions were being mentioned in social media, as well as measure how many people are potentially seeing these messages, whether good or bad.
Step 3: Establishing Benchmarks
The fourth and final step of developing SAS' plan included integrating web analytics data into the program to determine which programs were most successful at driving engagement and subsequent action.
Collection techniques were developed by testing and finalizing a master list of search terms that included all possible mentions of SAS and its peer competitive set. After experimenting with a number of different collection techniques, we selected Boardreader and eNR as the most comprehensive solution. We created a detailed set of coding and collection instructions that were approved internally at SAS. Next we collected data from traditional media and social media sources including: blogs, social bookmarking sites (e.g., Digg, Reddit, Farg, Newsvine), forums/message boards, micro-blogs (e.g., Twitter), photo sharing sites (e.g., Flickr), social networking sites (e.g., Facebook, Myspace, etc.) and video sharing sites (e.g., YouTube, Google Video, etc.). The third step of our plan involved standardizing and defining qualitative data for coding purposes. We wanted to define tone so we could know not just what people were saying about the organizations, but how they were saying it. We defined tone as follows:
• Positive-After reading the article, you are more likely to invest in, do business with, or work for the company.
• Neutral-The article either provides statements of fact (e.g., "xyz occurred"), doesn't give you enough information to feel either way, or it gives information that is both positive and negative, with neither side more convincing than the other.
• Negative-After reading the article, you are less likely to invest in, do business with, or work for the company.
In addition to tonality, we also characterized each item as either high visibility or low visibility depending upon where in the item the brand was mentioned. Tracking this metric would allow SAS to combine what people are saying, how they are saying it, and the likelihood of these mentions being read or seen by others. The idea behind this is that the more prominent a brand mention is in an item, the greater likelihood of a reader/viewer seeing and remembering it.
Additionally, we identified how each item positioned SAS and/or its competitors in relation to the industry or its place among competitors. Was it leading in the industry? Was it taking charge in a new initiative? Was it doing something that others have already done? We defined positioning as follows:
• Follower-The company was said to be following an industry trend, or taking actions already taken by its competitors.
• Laggard-The company was said to have failed to take an action that others have, and that has shown to be beneficial.
• Leader-The company was said to be a leader in its industry, or the first to take an action or make a business decision that may prove to be beneficial.
• No Positioning-The content does not discuss the company's position within the market, and does not compare it to its competitors.
We further identified what companies were mentioned, who authored the item and when, which spokespersons were quoted, what subjects were discussed, which business lines/industries were mentioned and how each item positioned SAS and/or its competitors on key issues. The selection of competitors to be tracked was made based on organizations SAS felt were their peers in specific marketplaces or competitors for particular offerings. Competitors were classified by one of two company types.
• List "A" included companies that only offer business analytics software and all mentions of the companies included in List "A" were analyzed.
• List "B" is comprised of companies that make business analytics software, but also sell data warehousing products, servers, operating systems or other types of software and/or hardware. Mentions of the companies included in List "B" were only analyzed when those mentions were specifically tied to their business analytics/business intelligence products as well as other products comparable to SAS to ensure an accurate competitive set.
Challenges and Obstacles
Establishing consistent collection methodologies was a challenge, particularly with Facebook and Social Bookmarking items. To deliver SAS with relevant findings, as well as methods that would be usable over time, we spent many hours developing coding categories for what we would track from media, such as Facebook and Social Bookmarking. Recording activity on these sites wasn't merely recording stories and related comments, such as in a typical blog. We had to understand and define entirely new terms such as walls, discussion threads, tags, and discussion boards, among others. Certain media types have different names for the same things (followers vs. friends vs. connections) while others offer options that are exclusive to that media type ("likes" on Facebook). Recognizing, grouping, and defining all of these metrics provided a large obstacle that was overcome through constant research and developing and testing definitions, categories, and coding results. Another obstacle was simply handling the large amounts of mentions found for each company. Social media has been gaining more and more popularity in recent years, which means working with thousands of mentions in various media. To overcome this issue we used a hybrid coding methodology inside our dashboard which allowed us to organize mass amounts of data and conduct data analysis across all variables.
With an ongoing view of SAS volume of mentions, spikes generated by product launches and coverage specifically of product launches, events, etc. becomes a telling metric by which SAS can estimate where, when, why and how its specific messages or news of certain products can be found in traditional and social media. SAS introduced a new business offering, SAS® Social Media Analytics, in April, 2010. Figure 14.1 shows the volume of SAS Social Media Analytics mentions across all media in the months immediately before, during and after the official launch date.
Figure 14.2 shows where the mentions occurred, emphasizing what media channel was most often responsible for the April, 2010 volume spike associated with the product launch. Because SAS Social media Analytics is tailored specifically to the Social Media realm, visibility of the product IN social Media was essential. As Figure 14.2 demonstrates, news in Mainstream media outlets accounted for just 2% of all mentions of the launch. A whopping 90% of relevant mentions occurred on Twitter amongst a fairly savvy group of Social Media enthusiasts. The remaining 8% of all SAS Social Media Analytics mentions occurred on additional Social Media channels.
SAS considers its positioning among its competitors in the media to be an indicator of success. It strives to not just have more mentions, but values quality over quantity. For them, success sometimes means more positive mentions and fewer negative mentions than its competitors; a larger share of positive coverage over time is often more important than earning the leading share of all coverage over time. Testing demonstrated that the sentiment of SAS mentions in comparison to its competitors was superior in terms of overall volume and positive mentions.
How do these test results specifically map to the objective(s) stated above?
Research Goal #1
Provide accurate and timely data to help SAS make better decisions. SAS now has a dashboard that it can access 24/7/365 to instantly get answers to its questions. Three members of the SAS team use it on a regular basis and as many as eight occasionally. They are able to monitor mentions of SAS as well as individual product lines specific to individual teams/departments within SAS communications. Team members can generate reports with the click of a button to share internally.
Research Goal #2
Correlate the data with online activity to determine levels of engagement and outcomes from specific programs SAS [and include] the ability to correlate online activity with media volume and the total opportunities to [display] SAS and/or a specific program or product line. By examining relationships between web analytics and media presence, SAS can demonstrate a link between the two in real time.
Research Goal #3
Advise SAS communicators on what they should or should not be doing in social media: What changes should they make to present programs, and/or what new programs should they add? The KDPaine team regularly dives into the SAS data, keeping a pulse on what is happening and continually looking for trends and changes occurring over time. The team is also acutely aware of many other trends within the technology industry, giving them an advantage when advising SAS on how to enhance their social media program. Regular consulting with the KDPaine team, as well as a data set that speaks for itself, gives SAS the information and drive to continually improve strategies and easily recognize success.
Research Goal #4
Set benchmarks against which future programs could be judged. KDPaine has been monitoring and measuring SAS mentions since 2004. With more than six years of data collected and analyzed in its dashboard, SAS has established an unlimited number of benchmarks for itself. Not only does SAS know at a glance where it stands in terms of media coverage amongst its competitors, it knows where it stands relative to yesterday, last year and five years ago. Every product launch, press release, special announcement, act of philanthropy over the last six years has been recorded. So, the media effects of every product launch, press release, special announcement and act of philanthropy can be measured against those recorded previously. Spikes in coverage over time are accounted for, whether they were the result of the inclusion of and gradual explosion in Twitter conversations or something that SAS did, like its SAS Social Media Analytics launch.
Review Questions for SAS Case Study
1. What challenges did SAS face when it decided to develop a tool to measure social media mentions?
2. What was the fundamental goal for SAS in developing their social media monitor tool?
3. How did SAS overcome the challenges and obstacles in measuring social media sentiments?
4. What were the results of the company's strategies for achieving its social media marketing goals?