Luke Elliott, Director Global Intelligence and Capability, Brown-Forman

Putting together a recipe for the perfect “dish” of data analytics can be challenging – especially when we expect people to cut their own slice, season to their own tastes, and choose their own flavor.

Kitchen metaphors aside, Synthesio can help data leaders prepare a nicely-structured serving that doesn’t require a brand’s marketing partners to consume the whole dish. After all, one person’s perfect plate is more important to them than knowing everything stored in the pantry. Those who enjoy the meal are often not those who prepared it.

At Brown-Forman Corporation, our social and web intelligence are just some components in a larger mix of digital and integrated marketing communications analytics. Internal stakeholders range from the marketing teams of our many brands, to more dedicated analytics-focused positions. This means the roles who benefit most from data are often not necessarily the people who are digital data experts.

Social media and online news/web data provides business leaders with a basis for context and comparison of the ebbs and flows of conversations, articles, reviews, and blogs across online landscapes. However, there may be a perception that accessing such data requires an advanced understanding of Boolean query logic. While the technical skills of an advanced team member will play a role to build the underlying structure of datasets, tools like Synthesio power the ability to ultimately prepare and serve up the core insights: creating intuitive, filtered reports that “mask” the underlying data queries and make it easier to present key knowledge to business leaders – without necessarily requiring data-intensive skillsets.

This functionality enables a more technically-savvy analyst to use advanced Boolean queries and data source filters to create the “nuts and bolts” that collect data and build visualizations – without requiring general users to understand or interact with the underlying functionality.

As an example, let’s say we’re interested in understanding cake flavors 🎂 – and differences in how people discuss different cake types.

We might create a basic query structure to capture mentions of a specific flavor in a very close proximity to the word “cake” (or “cakes”). We might also want to filter out some common “polluter terms” that add confusion in this type of query, like the words being used in association with nail colors or haircare — or any term that is sometimes related to our term but meaningless to the context of our desired analysis.

A basic Boolean query structure could look something like this:

(chocolate NEAR/2 cake*) NOT (nail* OR hair* OR salon*)

Then, we could repeat the same query structure as above for other flavors, or apply specific filters when we need a more streamlined result.

Let’s say we want to make some adjustments for cheesecake – such as being more specific with the proximity, and avoiding mentions of a popular restaurant:

(cheesecake* OR “cheese cake” OR “cheese cakes”) NOT (nail* OR hair* OR salon* OR factory)

Like any recipe, once we start building, things can be refined or expanded. Using this example, we could quickly end up with a very long string of Boolean text. But it’s neither necessary nor helpful to expect a general user to confidently view these query strings and instantly understand the implication and parameters. So…

With Synthesio, we can *mask* those underlying query structures – and create a simpler, clearer Topic/Subtopic segment that caters to a user’s specified data views and whatever filters they require. We can also apply some user-friendly colors that convey our data in relevant ways:

The tool’s ability to sort data, breakdown topics, and display relationships means end users can use Topics/Subtopics as a filter, without the need to navigate the underlying query structure.

In general, any user experience is well-served by this design ethos of “Click-to-analyze” while avoiding, whenever possible, a “Type-to-analyze” approach.

When it comes to identifying a desired set of data, every action required to get there is a barrier. Whenever possible, an analytics “baker” should focus their time on breaking down those barriers to success and minimize the risk of misinterpretation. The fewer barriers, the more likely it is that the tools we build will actually foster the organization to make decisions rooted in a broad –yet precise– view of data and the online landscape.

To help eliminate the “clickwork” required to mine the interesting discoveries, Synthesio can power your well-structured, segmented, pre-filtered layouts. Your users will benefit as you enable them to identify top insights without requiring significant input or action on their part.

Again using our cake example, we can develop segmented visualizations for key product landscapes, with pre-defined separate filters applied for each, like this:

Your end users can jump into topics of their own interest, which can lead to further context-based insights. And potentially, their own craving for cake. :-)

Of course, there are users who are more comfortable with the underlying query structure, and they might be inspired to take part in refining query builds. Your work might even awaken users’ desire to learn more about social & web data development, or grow their interest in query structures. For peers who are interested in becoming our “pastry chefs,” Synthesio offers different levels of access that system administrators can apply to unlock underlying builds, as needed. You can control who enters your kitchen, vs. who is simply enjoying the bakery case out front.

Maybe your business needs to understand and compare overall trends. Maybe you’re monitoring for emerging themes that may inspire a brand to take new action. Or, you might be diving deeper into the underlying context of specific topics. As a friend of data, you know that the less “work” is required to see key data points of interest, the more we can empower teams to allow data and insight to drive action.

Synthesio’s Topics-based query setup structure – combined with the ability to pre-filter and segment data for custom visuals – can provide the pan for the recipe you’ll put into your own oven to bake. All you need to do is determine the perfect recipe for your desired dish. Mix well.