Interview: MyHealthTeam’s Director of Research Beth Schneider on Using Brandwatch’s Data Upload API to Analyze 250,000 Data Points
By Gemma JoyceAug 13
Looking for deeper insights in your social data?
I’m often asked about this, so I thought I’d share some tricks I use in my daily work.
Read on for helpful tips on how to use Brandwatch to delve further into the data you’ve collected, and find those precious insights to help inform your strategies.
Without a doubt the number one secret weapon of all top Brandwatch analysts are the tools that allow you to divide the data up for more in-depth analysis – Categories and Rules.
By programming the platform to segment and graph the data in this way you can answer some of the deeper questions you or your clients may have – for instance relating to audience behaviors, or preference.
Once segmented, this data can also be filtered by Sentiment, Source and Demographic elements such as Gender, Interests and Profession.
Once created, Categories and Rules can be used future Queries too, as well as applied to other Brandwatch projects.
Where to find: In the ‘Tools’ section of your project.
Whilst Brandwatch defaults to tracking the volume of mentions this is normally useful only if you are looking to understand the total conversation around a topic.
The ‘Unique Authors’ option instead enables you to look at the specific number of people who have mentioned instead, giving a clearer perspective on the actual audience size.
Additionally, for those looking to understand the impact of your subjects Twitter activity the ‘Twitter Impressions’ can be very useful too, especially when investigating the value of social campaigns.
Where to find: In the Filters sidebar.
As you analyze individual mentions it is likely that you’ll find ones which will stand as good examples when demonstrating specific insights or trends.
I’d recommend creating a ‘watch list’ tag to assign these to at the start of your analysis so you can drop mentions into it on the fly.
This method can also be used to tag and exclude individual mentions which may contain spurious or irrelevant data.
Where to find: Once a mention is selected, in the sidebar to the right.
Whilst you’d normally focus on the number of individual mentions this approach can be difficult when it comes to comparing the percentage of elements such as sentiment or gender within multiple categories.
To help with this, Brandwatch allows you to chart your results by percentage, easily showing you the proportion of these.
Where to find: In the ‘View’ section of each graph.
It is fairly common when collecting and analyzing data to expect the results to match our pre-conceptions – a common behavior generally known as ‘confirmation bias’.
Whilst it is good to have a view, this approach can often blind analysts to the vital audience insights that a platform such as Brandwatch can provide.
In marketing especially, uncovering fresh trends can be the means to unlock new opportunities and markets – so if something looks a little odd or challenges your preconceptions go investigate it rather than disregarding or assuming it is an anomaly.
Social data is human data, and often human behavior can defy our preconceptions and have value when it does.