Benchmarking Social Data: The Key to Social Insights

Remember, you’re not analyzing data in a vacuum--it always needs context, and benchmarks provide the inflection point on which to build your insight stories over time.

Typically, when companies express an interest in brand health benchmarking, the discussion revolves around measuring a change in sentiment or share of voice over time. As they progress along the social maturity curve, their access to historical social data allows them to compare these metrics in subsets.

A retailer, for example, might want to benchmark key performance indicators from the last quarter of 2015 to compare them to the metrics generated by the coming holiday season. However, there is so much more to benchmarking social data, as it can be valuable beyond comparing your data against your competitors or your own historical performance.

Benchmarking and tracking the health of sub-brands is integral to gaining visibility into the volume and emotional impact properties are having on the overall enterprise.

It’s probably an oversimplification to say something like, “Every brand health benchmarking analysis is focused on three primary measurements: volume, time and sentiment.” However, at its core, it’s the truth. These three measurements can be sliced together to create dozens of useful and strategic KPIs.

A huge enterprise business like Amazon might not have worries about overall sentiment metrics or share of voice performance against competitors. However, Amazon’s reach extends beyond the core business of its namesake e-commerce website.

Amazon is the parent company of more than 20 internet brands. Some are niche e-commerce sites (Fabric.com), some are e-retail giants (Zappos) and some generate revenue through content and advertising (IMDB).

Benchmarking sub-brand share of voice and sentiment can provide insight into which properties are pulling their weight in driving consumer word of mouth and which might be having a negative impact on the enterprise as a whole.

What can benchmarking teach us? In looking at the charts below, it’s clear that there is cause for concern over the drastic drop of volume in mentions of Zappos. The final three months of 2015 generated 85 percent of the mentions about the brand in the nine months tracked.

Zappos is a significant revenue contributor, and a drop in word of mouth can translate into a drop in sales. Our benchmarking insight here is that perhaps it’s time to raise the ante on marketing and communications for Zappos–a company that built its reputation on innovative ways of reaching consumers.

PopeGuestPost1 PopeGuestPost2

The benchmarking of sentiment data at the end of 2015 allows us to generate insights from how sentiment trends over time. In the charts below, notice that despite the large drop in volume, Zappos’ sentiment remains constant.

However, the real story here is the shift in negative sentiment for both comiXology (12.8 percent) and Woot (10.5 percent). Without an initial measurement, the decline might not have been noticed, as it slowly creeped downward over time. Thanks to benchmarking, there is a baseline to return to for context around shifts in KPIs.

PopeGuestPost3 PopeGuestPost4

 

Another example of the power of benchmarking your social data can be seen with brands looking to leverage social data to assist in supply chain and inventory management, where a market research benchmarking study on a specific line of products can uncover a wealth of information that could send ripples across the organization.

A company like The Home Depot, for example, supports hundreds of thousands of SKUs (stock keeping units) of merchandise. The company’s e-commerce experience is a sophisticated study in best practices, as it provides a wealth of information designed to get the products consumers need into their hands, whether it’s in stores or online.

Ultimately, Home Depot would rather make a sale in a brick-and-mortar location, because there is a much higher chance that the consumer will consider and purchase additional items in their path through the store. So the website features on-site inventory, actual aisle and bin location and pre-purchase pickup options.

Maintaining a store inventory of hammers and nails doesn’t make much of an impact on the location’s revenue generated per square foot. But when it comes to much larger items, companies like Home Depot struggle with maintaining a brick-and-mortar inventory sufficient to meet client demand.

For example, August is typically considered the best month of the year to purchase a snow blower, as retail stores look to clear the remnants of last year’s inventory to make room for new product. The problem here is that last year’s remnants are frequently the most subpar products in the store. There’s a reason why last year’s in-stock snow blowers didn’t sell–and it’s probably not a good reason.

Home Depot has limited in-store space to support large inventory items. Its capacity to maintain an online inventory is much greater, as it can rely on local partners and drop-shippers to meet demand.

Determining inventory allocation for something as large as a gross of snow blowers is a major undertaking that can’t rely on historical sales figures, specifically because not all products are offered in the same regions, channels and outlets. Beyond looking at last year’s sales figures, how can a company like Home Depot determine which snow blowers to invest in for in-store promotions, and which to offer solely online?

A benchmarking analysis of conversations around specific brands for sale in a retail establishment can help determine where inventory should be allocated for the coming winter season. Obviously, using existing correlative inventory data would provide context to the analysis, but simply benchmarking sentiment for a product line, like snow blowers, before a major event, like the Northeastern U.S. blizzard of 2016, could provide major insight into what products you should be stocking for the coming winter.

The baseline for consumer sentiment around Home Depot’s three primary brands of snow blowers was 98 percent positive versus 2 percent negative on average. Take a look at the post-storm sentiment:

PopeGuestPost5

It doesn’t take an expert in heavy machinery to see that Ariens snow blowers caused consumers some major issues in their attempts to dig out from the storm. Data like this makes inventory management a breeze. It’s difficult enough to make space for such huge equipment, but even more difficult to accept it for return. The right bet here is to take a pass on Ariens–and take a gamble on Toro.

As you can see, garnering insights from benchmarking takes patience and a little creativity, but it’s integral to any social media listening strategy’s ability to truly pay off over time.

Remember, you’re not analyzing data in a vacuum–it always needs context, and benchmarks provide the inflection point on which to build your insight stories over time.

Leah Pope is chief marketing officer of social insights provider Synthesio.