Brandwatch in 2019: A Year in Review
By Leia ReidDec 30 2019
Tomorrow, users of Brandwatch will see the brand spanking new Query builder in all its glory within the platform, as well as our new Sample Queries functionality. We’re a bit bonkers about Queries here, and we know many of you are too, so we’re very excited.
We can’t go on enough about the importance of good Queries. Queries are the key to good data, and good data is the start of good analysis, and good analysis is…well, you get the idea.
The fact is, having a super flexible Query builder like ours means better Queries and better data – something that’s simply not possible by using a handful of basic booleans or just searching for a singular keyword.
But we also know that with flexibility can come complexity.
22 operators can be a lot to get your head around if you’re new to Query writing, but once you’ve mastered them they will make you feel like, well, a Query master. See the bottom of this post for lots of resources to help you get there.
With all of that in mind, we wanted to make the process of writing Queries easier, more intuitive and more fun all round, so that our users can get going with their super useful and relevant social data.
We wanted to rethink the Query builder from scratch – it hadn’t changed in a couple of years and is the start of most people’s journey in Brandwatch, so it was time for it to get some development love.
We always take user feedback very seriously, so we did user testing throughout the development of the new builder and spoke to some of our heavy Query writers to understand what would make writing Queries easier.
“Brandwatch has taken what is a necessarily complex feature and made it more user-friendly without compromising utility. As always, the Brandwatch team has carried this out with a passionate commitment to feedback-driven UX.”
– Elliot Darvick, Social Strategist, Spark44 Los Angeles
What resulted was a whole load of changes, from improving error messages to making the experience more friendly:
Better error messaging
We’ve always had error messaging within the Query builder, to let you know when something is stopping the Query from working. We’ve further refined these to make them more helpful, with a range of added features to help you correct your Query if you do make a mistake:
We found many users were saving their Query, then going to the Dashboard to look at the data, then going back to their Query to refine it. We wanted to cut this additional step out by allowing a better preview of the data in the builder.
The first step is adding in a clickable history chart that shows the volume of your Query results over the past 7 days. Not only does this give you a better idea of volumes, but it also allows you to click in and see which mentions drove particular spikes, so that you can refine your Query if you find irrelevant data (e.g. a big spike caused by retweets of a spam tweet).
We’re intending to continue adding in more preview tabs to reflect components within the Dashboards – watch this space!
We’ve made the experience of writing Queries better all round by making the Query builder nicer to use. That includes small changes that make a big difference, such as:
Plus, as well as all that, we’ve introduced the ability to run Sample Queries, opening up new types of analysis where just a percentage of the mentions is needed – read more about that here.
And this isn’t the end. The Query builder is going to continue being developed, so over the coming weeks and months you’ll see more and more new features added to help you write and refine your Queries.
Brandwatch users can find out more about the changes, including a video guide, here.
As part of these changes, we decided to remove the ‘Structured’ Query option. In the old Query builder, you could write a basic Query by inputting terms into three boxes: inclusions, exclusions and context.
Whilst on the face of it this seems like a great, easy way to write Queries, we in fact found that it was incredibly difficult to get the right level of refinement using this method – meaning it was much harder to get relevant data.
It was fine for a unique brand name like Sainsbury’s; less so for a more ambiguous term like Gap.
We want our users to get the best data possible – and the structured Query Builder wasn’t encouraging that. Our users obviously agreed, as very few of them were using this option, so we’ve removed it.
No problem. We’ve got tons of help guides, blogs and videos, including all the great stuff in the new Brandwatch Academy:
You can also find a selection of Query resources right here on this blog – there are some useful resources here, including our boolean operator guide, and you might also want to take a look at:
For all blog content about Queries, take a look here.
If you have feedback on the new Query Builder or are longing for some Query content we don’t already have, let us know!