Finding Meaning In Billions of Voices: A Look Under the Hood at Brandwatch

The prevalence of data in every aspect of our lives means that it’s now possible to quantify almost every aspect of human interaction. “Soft” metrics like loyalty, popularity and reputation can now be represented in numbers drawn from raw data, information that comes from sources with which we interact daily. Social platforms, forum comments, review sites, and at a deeper level, metadata from media and interactions with applications — all these create patterns in behaviour and expressions of opinion that are hugely beneficial to organizations concerned about their reputation, and present and future activities.

The first marcomms platforms focused on limited sets of inputs, like pulling follower counts from Twitter or accumulating “likes” from Facebook. But as technology became more ubiquitous in the daily lives of much of the planet’s population, these tools’ limitations have meant that businesses now look to more data-focused platforms to derive empirical metrics from an extensive range of sources numbering in the millions.

Unfortunately, while increasing the number of data sources is an excellent concept statistically (the more information gathered, the more accurate the resulting figures), a greater number of sources means a greater chance of incompatible data formats, data duplication, and data corruption.

Getting a complete overview of consumer interactions, needs, and opinions takes a very specific combination of applied marketing technology and the types of tools that would be more commonly found in the silicon toolbox of a data scientist — the new breed of professional versed in statistical methodologies and steeped in data parsing techniques.

Even across a handful of social platforms, it’s easy to lose track of what the collected data offers amid a snowstorm of mentions, likes and re-posts. With many consumers (and business decision-makers) using the internet for everything from casual chat to choosing potential suppliers, it’s important to  be able to unearth core messages, trends and prevailing thinking.

Drawing together the disparate threads of opinion from right across the almost limitless scope of the internet on a brand or product is a tough call from a purely data-based perspective. The only platform we at Tech HQ are aware of that’s capable of doing so is Brandwatch, a capability it combines with AI-powered, language- and sector-aware sentiment classifiers. Because of its capabilities under the hood (more on these anon), it can act as a canonical source of sentiment analysis and digital consumer intelligence for companies and brands. It searches for relevant data from multiple sources and collates information, de-duplicating and sifting on the fly to dig out inconsistent or mismatched data formats.

Users can drill into different market segments according to desired outcome, be that competitor analysis, grabbing feedback to inform new or existing products, discovering untapped markets, or finding out why specific products are flying off the shelves here, but are gathering dust here.

Insights and any results can be presented in various ways — headline figures suitable for C-Suite consumption or granular metrics for research experts or Marketing’s frontline troops.

The multitudes of online conversations held on different platforms can expose strongly-held consumer opinions in a specific market sector, like cosmetics, for instance. Once gathered and distilled, opinions can – and perhaps should – inform companies’ policies. The Body Shop’s Jennifer Rice, Director of Customer Strategy and Analytics noted that there was a lot of online chatter about the launch of their Bond Street store’s new format. Brandwatch surfaced key insights, discovering that the introduction of a refill station dominated 90% of the conversation: “It was a lightbulb moment for us,” Jennifer explains. “Really what it did was give us the confidence that refills are what we should be doing and helped us accelerate [our] program.”

Brandwatch’s “secret sauce” is its AI engine that intelligently sifts the information that flows across the platform, drawing out the insights that might escape human operators’ notice. Using technology in this way makes perfect sense, given that the Brandwatch platform’s data records go back to 2010, comprising billions of conversations from sources like Instagram, Facebook, and Twitter. Add to that, a roll call of the millions of other websites out there, some of which that might be more relevant to a brand, for instance, than a single social platform, like Reddit. Given the numbers involved, it’s quickly apparent that a thousand human brand analysts would struggle to keep track of even the broadest trend.

The Brandwatch platform goes further: companies can plug-in their own data, collected over the years and held in other platforms, online or offline databases, even archives of old Excel sheets. Those are sources of valuable information that generally would need significant treatment to be manually imported. Brandwatch makes it easy to bring in and have the AI engine surface the inherent value.

Like The Body Shop example above, real-time data is increasingly important for many companies that need to see opinion metrics as campaigns are rolled out, products are launched, or crises emerge. Thanks to Brandwatch’s direct connections to major social platforms and websites, that’s possible, with practically no lag between expressions of opinion online and a company seeing aggregated results in their Brandwatch dashboards.

While there’s a great deal of data verification and analysis going on under the hood of Brandwatch, the company also publishes details for API access to the platform. That means any other real-time  data sources can continue to pay dividends.

Even without this extensibility, the platform draws its information from literally billions of sources comprising hundreds of millions of websites and social platforms; scanning images and text to track brand health, campaign performance, public perceptions, competitors’ performance, online behavior, plus past, current, and emerging trends. It surfaces insights to multiple stakeholders and helps distribute results in attractive and insightful formats.

Brandwatch’s list of 2,000+ clients reads like a who’s-who of the world’s biggest brands. Among them are major B2C retailers and financial service providers, whose public image is a critical part of their success. There are also many B2B companies that rely on Brandwatch to help uncover broader public perceptions and trends to help them stay relevant and grow their brand reputation.

Regardless of business model, sector or company size, no organization is immune from public expectations for companies to create a more sustainable and fair world while simultaneously providing excellent customer experiences.

A brand’s success depends on its ability to adapt, anticipate the next opportunity or forsee  looming risks before its competition does. Brands like The Body Shop use Brandwatch’s powerful insights to analyze and quantify the nuances of human emotions expressed about and around their products, helping them find what’s popular now, what’s starting to trend, and even what tomorrow’s go-to products might be. Your organization and brand(s) can join them — reach out to a local Brandwatch representative to discuss your options.