On Tuesday, I had the pleasure of attending a social media measurement session led by VP InLine Analytics Margot Sinclair Savell (@margotsavell), one of my favorite Weber Shandwickians (yes, that’s a real term in Weber World). Speaking to a full house, she didn’t disappoint.

Presented by the Puget Sound chapter of the Public Relations Society of America (@PRSAPugetSound), Margot’s session, How to Measure Social Media: Your Brand’s Influence, walked its audience from start to finish on how to use a “social media snapshot,” or brand measurement strategy for social media, to create and measure effective campaigns.

To avoid penning a massive blog-essay to cover all of the event’s great content, I decided only to highlight what I considered the key takeaways from the talk.

The Media Landscape Continues to Change

With the ongoing fragmentation of media sources, deteriorating media economics and insatiable demand for content, change continues to rock the media world. And this fast-moving change can prove to be an overwhelming obstacle for brands looking to measure their online footprint.

Margot drew an intriguing comparison between the current social media landscape and Tokyo’s complicated train system – and I couldn’t agree more with the analogy, having firsthand experience helping clients sort through seemingly endless streams of data and getting lost in Toyko last year after jumping on one wrong train after another. In both scenarios, clear guidance is needed if you expect any form of desired results.

Set Goals, Match them to Tangible Objectives

As with any strategy, tactic or campaign, measurement efforts should always be built upon clear goals that are tied to measurable objectives. Metrics, timeframe and all other aspects of a measurement framework should grow from clear, predetermined goals and objectives.

Here’s an example – Goal: Improve brand sentiment. | Objective: Increase the percentage of positive online brand conversations by X percent. Seems like a no-brainer, but essential nonetheless.

Look Beyond the Numbers!

Fellow colleague Elaine Norton (@ehelm) tweeted a quote from Margot during the event, “Data without analysis is meaningless.” This is true. Too often brands get caught up with collecting every quantitative metric possible and end up neglecting a deeper, qualitative dive into the data. Although hard numbers are undoubtedly important, they lose their effectiveness to inform strategy if they aren’t supplemented by meaningful context.

Margot brought up a great example of this. Compare Twitter User A with 10 followers and Twitter User B with 10,000. First instinct is to dub Twitter User B as more influential. But if you look past the raw stat of follower count, you may find that Twitter User A’s 10 followers all have 5,000 followers of their own, while Twitter User B’s 10,000 followers are comprised mainly of non-engaged bots and “follow-me-and-I’ll-follow-you-back” drones. A look beyond the raw numbers can literally flip interpreted findings on their heads.

Measurement is an Ongoing Process

My last takeaway relates to the measurement lifecycle. Measurement should be an ongoing process, not something to be done exclusively at the end of a campaign. Effective campaign measurement requires measurement at the start to create a benchmark, then ongoing, consistent follow-up measurement to provide insight into tactic effectiveness and progress on goals and objectives. Continual measurement can provide ongoing strategic guidance and retroactive analysis versus end-campaign measurement which only provides the latter.

That about sums it up. Judging by the great feedback we’ve been receiving, it feels that the audience’s measurement knowledge increased tenfold from Margot’s talk. If you want to build your measurement know-how too, be sure to follow Margot on Twitter and definitely take a peek at her four-part measurement blog series.

If you also attended the event, please feel free to share your key takeaways in our comments!

Leave a reply

Your email address will not be published. Required fields are marked *

Go top