Last week I was able to present at the Obtaining Value and Insight With Social Medial Analysis event hosted by IBM Analytics Solution Center. The abstract for the event was:
The internet is in a second wave of innovation. At the turn of the millennium, we saw a technology-driven, website-centric new-world-order. We are now part of a new wave of relationship-driven, internet-based insights.
Organizations that have the capability to harness social chatter and analyze it to provide intelligent insight will benefit greatly from this revolution. Regardless of whether your organization is a government agency, an academic institution, a health care provider or a private sector business, There is much to be gained from adding social media analysis to its “toolkit” of enabling technologies.
You can see my slides and also the other panelist slides by following this link. There are also some audio files as well, so hopefully you’ll check out the presentations. Other presentations where:
- Extracting Meaning From Social Media With CCI, Edward Burek, IBM Software Group, Application Architect
- Future of Social Media Analytics, Rick Lawrence, IBM Research
Both speakers where fantastic, and provided some really amazing insights to the work that IBM is doing in the world of social media analysis. The tools that IBM have developed are remarkable, and help agencies measure the impact of their social media initiatives. I’d really encourage you to take a look at their presentations.
Throughout the event, I thought there were a lot of really interesting lessons learned. My presentation focused predominantly on how agencies can start to discover ROI from their social media program, and use data to drive improved decisions. I started off by posing some basic questions about ROI for the public sector:
- What are we trying to quantify?
- What impact are we measuring?
- Who is our core audience?
- How does this ROI fit into our agency mission?
- What do I do with all this data? Where can I find knowledge?
The core of my presentation was identifying a process that agencies can start with to understand ROI. During the presentation, I mentioned that agencies should think about what the process is for how they evaluate other programs, and apply that model to social media. From this, I gave the attendees a 10-step model to think through about measuring social media. Like a lot of strategic planning and models like the one below, one the key benefits is investing the time in the process, and thinking critically about the initiative. It’s a great first step. The steps I outlined were:
- Define Scope – Why are we doing this?
- How will you define success? – Think Traditionally
- Consider how you can design your evaluation
- Collect Data
- Show Impact and Quick Wins
- Analyze
- Share
- Feedback from Stakeholders
- Share Your Success, Learn from Experiences
- Improve and Repeat
My second point is that agencies should work to develop a scorecard for their social media plan. This is just a quick spot where social media managers can write down key metrics, track trends, and understand what is working. With this kind of data quickly on hand, they can share success across the agency and hopefully build some support around their social media initiative.
Another point that I talked through was if agencies can actually derive and ROI, and mentioned quickly if in the public sector if it is Return on Investment, or Return on Influence. In my opinion, it is likely both – but the challenge for agencies is that defining what constitutes “investment” and “influence,” which is extremely complex and challenging. Again, the trick is to think in traditional terms. If you want to measure investment, consider cost savings from social media.
In order to do this, social needs to be tied to mission centric initiatives, and will be one component of a strategy. For instance, if you would like more citizens to submit online forms, you could use social media to promote the service and track how many forms have actually been submitted online. You can track how people are getting to your site, using referrals and if Facebook is one of the primary drivers, you can add more focus to that strategy. The cost can then be calculated by looking at how the new online process reduces costs from paper forms.
Influence gets tricky. There are many ways and services available to measure influence. One way I think about influence is in terms of reach and based on recognition. If your agency is seen as a thought leader and the “go-to” resources on the web, then your influence will be high. Getting at influence is tricky to me, and I would love to hear how your agency defines influence. I feel like this topic warrants itself to its own blog post.
All in all, it was a great session, and I learned a lot about how IBM is leading the charge in social media analysis, and it was great to share some of my insights and some of the common challenges for social media analysis.
 |
The IBM Analytics Solution Center (ASC) is part of a network of global analytics centers that provides clients with the analytics expertise to help them solve their toughest business problems. Check out their Analytics to Outcomes group on GovLoop. |
How to measure influence? It’s probably hard to do in the abstract so let’s think of a specific mission. How about FEMA? So say FEMA wants to launch a campaign on civilian emergency preparedness. And part of being prepared — let’s say for an earthquake — is to buy specific disaster supplies. So in addition to your traditional metrics like #retweets, you could analyze national economic data to see if there’s an increase in purchases of the recommended disaster supplies. So this is a pretty simple an example, and I am sure there’d need to be some fancy normalization of economic data, but I think you get the idea.