2011 has been a year of extraordinary growth in our practice at Semphonic. But no part of that practice has grown as dramatically as Social Media Measurement. There's tremendous interest right now in Social Media but it's an area - perhaps even more than mobile - where companies are still struggling to formulate a coherent strategy. That hasn't necessarily stopped or even slowed the pace of investment and maybe that's not unreasonable. If you wait until everyone figures out how to do this stuff, you may find that your competitors enjoy insurmountable advantages.
With social media measurement, it often seems to us that it's a case of "deja vu all over again." Companies make the same mistakes they made with the Web and Web Analytics. Before you're a victim...STOP...remember these few simple principles we've learned about effective Digital Measurement:
1. Develop a Plan
2. Tie Measurement to the Plan
3. Centralize and Standardize before you Democratize
4. Abstract Collection from Analysis
5. Pick the Appropriate Tools for Each Distinct Job
They apply to Social Media Measurement too.
Over the course of a few posts, I'm going to cover some of the techniques we've been creating to do Social Measurement and talk about some of the cooler analysis projects. Right now my list of topics includes: measuring influencers by multi-topic relevance, tactics around keyword profile building, automated translation methods for internationalizing profiles effectively, creating time-trended word-clouds, automated Excel integration for Social Dashboards, breaking out social measurement by source channel, replicating conjoint product analysis with social verbatims, measuring quality of engagement by social campaign source, measuring the value of a Facebook Fan (or similar relationship), plus some broader topics around the integration of Social Media into Web Analytics and into the type of Customer Intelligence System that I mentioned in my last post. I also hope to touch on several very interesting online survey projects we've been involved with and talk a little bit about what I see as a fairly dramatic shift in that particular space.
Hmm. Maybe it's more than a few posts. That's a lot of stuff, though no doubt some of it will drop by the wayside. Still, it reflects the diversity of social media measurement tasks that we've been involved with this year and that, as it happens, brings me to my topic today - the many uses of social media and the corresponding measurement disciplines.
In 2011, we've seen our clients use social media in at least six very different ways:
For each of these, there is a corresponding set of measurement techniques and tools.
For organizations that talk about "social media" as it if were a single capability, this view is often a rude shock. It's import is clear. One of the first decisions when you talk about a Social Media Strategy is deciding which of these social media functions you're interested in. Not only that, but the type of measurement and the necessary tools for each function are quite different. Here's another rude shock. There is no single social media measurement tool that even comes close to meeting the full range of needs if you are tackling all or even most of these functions. In general, it's fair to say that even the most versatile (which isn't to say the best) of social media tools only support two or perhaps three of these functions. There is also a distinction between tools that can provide for the collection of data (the listening) and tools than can analyze the data. Many tools try to combine both functions, but we've found that it's often much better to disassociate them.
We tend to think of Social Media tools breaking out like this:
There are distinct Platform tools for Customer Support, Public Relations, Social Campaigns, and Communities. Social CRM and Product and Customer Research tend to live in existing tools or in measurement tools. The tools in Collect & Report support the Products & Customer Research, Social CRM, Campaigns, and Public Relations functions. The API Aggregators tend to support Communities, Social CRM and Product & Customer Research. The Analytics Systems are primarily focused on Product & Customer Research.
As a measurement company, we tend to be concerned with the last three categories more than with the Platform tools themselves. However, it's been our (and our client's) experience, that when companies try and blend these (by getting a Platform Tool that's also meant to be a Reporting or Analytics System) that the result is not pretty. On the other hand, we've also found that the Platform Tools are the only reliable source for operational metrics and are, generally, the right "system of record" when operational metrics are necessary.
I know some of my categories are not exactly industry-standard, so I've taken the liberty in this next graphic of showing some of the representative tools in each of the three categories where we are directly concerned:
When you carve up tools this way, I think it makes it much easier to understand what functionality is important to you within the tool set and which tools play together and which are competitive.
It would be natural to pair up Clarabridge with Next Analytics and NM Incite. But if you're pairing up Radian6 and NM Incite, then you're probably duplicating your efforts.
I started this post with five process steps learned from Web analytics about how to do good digital measurement.
Translating those steps into the world of social media measurement means understanding which of the six (or other) social functions you're actually trying to tackle. It means realizing that for each of these functions, there is a distinct set of measurements (and, no, it's not Brand Mentions). It means that for you to get it right in your organization, you'll need to centralize those measurements and standardize them before you can democratize them. It means that you shouldn't confuse collection with analysis and assume that the two must go hand-in-hand within a tool. Finally, it means that there isn't one tool for managing Social Media and there isn't one tool for measuring it. You can't expect one tool to be useful for advanced textual analysis (like Clarabridge or SAS) while also supporting your PR workflow. It's not going to happen.
I hope this provides some broad context on the range of social media tasks, the related measurement functions, and the tools appropriate to each. Next week (unless I get derailed by eMetric's posts) I'm going to touch on the work involved in building out a complete social media measurement program and show some samples. After that, I'll dive down into the detailed topics I mentioned above.
[If you're out at eMetrics, please come by on Wednesday (3:30PM) for my panel on Sentiment Analysis or on Friday at the IMC (1:30PM) for my presentation on Centralizing Measurement and Managing the Agency Relationship. Hope to see you there!]