In recent years, investment in data analytics has been on the rise, with the International Data Corporation (IDC) projecting a worldwide market for big data and business analytics solutions that exceeds $274B by 2022. Until now, collecting and connecting internal data has been the primary focus; building connections is often the hardest part, as different teams built their own systems over the years that don’t use the same logic, language, or platform. Expanding an analytics strategy to include external data is exponentially more difficult, but it’s the necessary next step for brands that want to stay competitive.
Why internal data isn’t enough
While important, internal data doesn’t tell a complete story. For one, it’s backward-looking; sales figures, for example, only tell you how a product sold in the past, not how it will sell in the future. Internal data is also narrowly focused on a single brand or organization. For a broader view of the whole market, your brand’s potential, and all consumers (not just your customers), you must have access to external data as well.
But collecting and connecting external data is a far bigger initiative than most brands can tackle on their own, for two reasons: the sheer volume of information available and the complete lack of control over how data is organized. Every day, the internet adds 2.5 quintillion (billion billion) bytes of data, and more than 80% of that data is in totally unstructured and disparate forms. Somewhere in that data is information that could be transformative to your business, like:
External data offers many angles or views on the same topic, creating a holistic view of the entire market. But tapping into this data and generating valuable insights from it requires the kind of advanced artificial intelligence (AI) capabilities that most organizations lack.
The solution: The right analytics partner
There are plenty of firms out there that claim to offer robust external data analytics. But many of these platforms are limited to pulling in data from one particular data source or data type.
Signals Analytics has broken the mold by solving for the complexity of extracting large and disparate datasets and extracting high-quality information from that data. We offer all this information on a configurable platform that generates tailored insights for different industries and business needs. No other data analytics platform collects and connects data the way we do.
How we collect data
The first step to generating our revolutionary market intelligence is researching and choosing appropriate data sources for a given market. Data sources are widely varied across three main classes: third-party API integrations with data suppliers, internal data sources, and publicly accessible websites. Maintaining continuous access to those sources at scale is a complex engineering challenge that requires constant code writing and updating.
While internal and external ecommerce data provides a view into the competitive landscape, social conversations are key to understanding consumer sentiment. Typical social listening tools will track a single brand and perhaps a few competitors. But this data will be very limited. In contrast, Signals Analytics will track all social activity relevant to a market. So instead of following specific yogurt brands like Dannon and Yoplait, for example, we’ll track all mentions of the yogurt category regardless of what brand is or is not associated with them. We capture similarly broad keywords in all other data sources, from specialist sources to sales data to demographic information, in order to widen the lens, present data about the category as a whole, and expose missed opportunities.
Data collection has its challenges, but it’s actually the easiest part of our process. The way we connect data, on the other hand, is completely unique and incredibly powerful.
How we connect data
Some data sets can be connected to each other easily via universal codes, such as UPCs or SKUs. But this type of identifier is only available for a few specific data sets. In order to connect patent filings (which never include SKUs) with products, for example, we need to rely on something else: the source content that can be extracted and connected to other data sets.
Our proprietary machine learning and natural language processing (NLP) algorithms extract content from unstructured data with a high level of accuracy. Normalizing that content allows us to ensure that all data sets are translated into the same language, using the same taxonomy and values so that each data set can be compared to every other.
Others in the data analytics space are limited in their ability to make these kinds of connections, so they either don’t make any connections at all and consider each data set separately or they make connections between much smaller sets that result in limited insights. The connection nodes we build span across every data set we have, and we do it at scale.
The impact of connected data
With fully connected data, the insights that can be gleaned are staggering. Through the Signals Analytics dashboard, you can choose a topic, feature, or benefit - let’s say, yogurt, flavor, or probiotic content - and drill down into every related piece of data, from consumer discussions to research papers, to achieve a complete understanding. Alternatively, you can dive into a specific type of data - online discussions, for instance - and compare consumer discussions vs. expert discussions to see if they are aligned, revealing opportunities for better consumer education. The diversity of our data connections also makes it possible for us to offer predictive dashboards like our Customer Needs Assessment. This dashboard presents the products available in a market ecosystem alongside the consumer discussions about that ecosystem to expose any gaps between the two. With this information, a brand can proceed with new product development, improved product benefit messaging, or untapped marketing opportunities with confidence.
The next step: contextualization
Collecting and connecting data is just the start. Contextualizing that data is key to generating real insights; for more on that topic, check out this blog post. To learn more about our entire process, schedule a demo or download our “Under the Hood” white paper.