Perhaps you’ve noticed certain buzzwords that attempt to define a changing landscape in digital analytics: “big data,” “mobile marketing,” “predictive analytics,” and even a term I coined here in 2012: “convergence analytics.”
Convergence analytics encompasses all of the emerging trends, and you can download the free report about that here.
But somehow you have to navigate the next couple of years as new measurement paradigms mature and become mainstream. Understanding what’s behind the buzz can help.
First, let’s deconstruct some of the buzzwords – it tends to make them less scary. Then after each definition, we’ll add a few notes as to what you can do now to take advantage of what’s really going on.
Buzzwords and What to Do
1. Big data. There really is no such thing. It’s a term used by business journalists and product marketers to create fear and doubt for the digital marketer. Yes, lots of data is being collected. Yes, there are tools that now can analyze multiple streams of data. Yes, you should try to make decisions based on data. But “big data” is not an enterprise in and of itself. It’s just a word for “lots of data” and some of the practices evolving as folks try to utilize that data.
What you should do. Find out what kind of data is being collected within your organization, and how much of that might affect marketing (in a broad sense). Decide what data might help you make decisions and take a look at its current form. Determine if you can use it as-is, or if you need to pull it into a tool you a) already have or b) need to subscribe to that will help you make decisions. The keyword is “decisions.” If you can’t use the data to make any decisions, leave it alone.
2. Mobile marketing. Ask five experts about this and you’ll get five different answers. No one really knows if this is a category yet, or several categories, or what the rules are, or how to win at it. Again, application vendors enjoy talking about “mobile marketing” because they want to sell a tool that some marketers may use to “optimize” their “mobile marketing.” But we haven’t even figured out what success looks like, and on what devices. And if you don’t have a mobile strategy yet, you’re not alone.
What you should do. Decide what platforms matter to you. Do you need to reach people on their smartphones? Or maybe just their tablets? Or both? Do you need an app to deliver the full value of your product or service? Or do you want to deploy adaptive screen technology (e.g., your site looks different depending on the device being used to view it)? Are you advertising to people on their smartphones? Or not? Does your product or service gain any particular relevance by being specially constructed for mobile users? And most importantly, where do your customers convert: on a device or in a browser? If you can start answering these questions, you will have the foundations for what is commonly called “mobile marketing.” The healthy approach is to go through some of the above exercises and then decide if you need a mobile strategy at all. Depending on your business, you may, or you may not!
3. Predictive analytics. It sounds like you should now expect to see into the future and ensure outcomes. Except not. Predictive analytics is essentially a data modeling exercise that looks at historical trends related mostly to campaign ROI and then allows you to review them in a dashboard. Then, in the same dashboard (generally), you can change some of the parameters – chiefly, how much you would spend on a campaign – and then allow the data model to extrapolate that out into the future. So, if you then advance the date range of your report into the future, you can see what your “future ROI” might look like. Often, these predictions turn out to be fairly accurate but there’s certainly no guarantee.
What you should do. If you have a large marketing group that does lots of campaigning, you should certainly be tracking campaign success and trying to determine best outcomes for future campaigns. If you want to automate this somewhat, look for the predictive capabilities already built into your measurement tools. Many digital analytics tools today have some predictive capabilities in them, and often these are overlooked. You may need some help from the vendor or an outside expert getting these capabilities implemented, but the insights can be very useful when properly modeled. Hint: this capability is more commonly included in the existing tools on the market than you might think (because it’s not very complex from a technology point of view).
4 . Convergence analytics. This term defines the confluence of technologies like: multi-channel data collection, powerful algorithms, and sophisticated display layers. The results can be powerful – a little bit like what used to be called “business intelligence” – but with a kinder, gentler approach. And geared specifically for marketers. There are at least 50 and probably more than 50 vendors converging on this space right now: all laying claim to the ability to connect lots of disparate data sources and put them into one dashboard or a series of dashboards.
What you should do. Accept the multi-channel future. Web analytics or any analytics in a silo just won’t make hay anymore. The new paradigm is to mash data together and find correlations. The warning is: beware of false positives. Just because trends look similar doesn’t mean there’s causation between one and another. But there’s no doubt that looking at multi-channel data in an aggregated form can lead to rich insights.
Image on home page via Shutterstock.