Introducing Audience Uploads
By Mercedes Lois BullNov 18
We know, we bang on about user feedback a lot.
It’s true, but it’s because we really do value it and it has a huge impact on our product development.
We take your feedback into account every step of the way and in every area of our business, but of course it all starts with the roadmap – what we are actually going to build.
Our space moves very fast, which means that, although we have a long term ‘mission’ and a mid-term feel of where we’re headed, our actual roadmap rarely looks past the next 3-6 months; doing so would be pointless as our clients’ and the market’s needs change quickly and we’d soon be behind.
When developing our roadmap, a key source for feature ideas, among other things, are requests from clients and staff.
All feature requests from clients and our team are put into a system we use called ProdPad – there are hundreds of ideas in there at the moment, each of which can then be upvoted by our staff when another client requests it or if they think it’s a particularly good idea.
We evaluate all of these requests on an ongoing basis (I really do read every single one!), and a large number of them make it onto the roadmap.
A feature request in ProdPad
Of course, we can’t possibly fulfill every request, but requests that do get implemented generally fall into one or more of the following: it will benefit a large number of clients, it is requested often by multiple clients, and it aligns with our overall strategy.
Sometimes, if a feature request is small and easy to develop it’ll also get onto the roadmap, even if it is a bit more niche (a ‘quick win’!). An example of this is the search box we added when users are sharing Dashboards.
Requests never get totally forgotten – we will revisit ideas if they become more feasible at another time.
Some of our biggest features, both launched and in development, came out of user demand – as well as many of the smaller improvements we regularly make, including:
That’s not all – some super cool reporting and alerting features we’re currently working on also came out of user demand.
And it doesn’t stop there. Even once we’ve decided what we’re building, we continue to incorporate user feedback into our development.
We work in an agile way, meaning we try to get MVPs and mock ups in the hands of real users as early as possible, and then use their feedback to develop the next iteration.
We do this through our beta programme and user testing. Our beta programme is comprised of volunteers and key users who’ve been identified as interested in a feature, who get early access to new features in return for feedback on what works and what could be improved, helping us understand if our assumptions were right or if we’ve got something wrong.
This is coupled with user testing – either via phone or face-to-face – by one of our development team, where we can observe how a user interacts with a new feature (either live in the app or using interactive mock ups using Axure). It’s an incredibly valuable way to understand user behaviour and see where the holes or stumbling blocks are in a feature.
With all of that feedback and insight, we continue developing the feature, changing our designs as we go.
Often, we were on the right track, but sometimes we find that we need to rethink some of our decisions. This was the case with Facebook Channels, which changed significantly from its early incarnations due to feedback in the beta period.
Before (above) and after (below). We went from one ‘mega’ component to splitting this out and creating a whole new dashboard, among other things. And check out that 4-axes chart!
Feedback also played a big part in the development of our new Query builder, launched this month, as we gathered feedback before and during the process of redesigning it, taking into account how people use it when writing Queries (or not, as was the case with Structured Queries!).
Then, once all is done and a feature goes live in the app, we continue to take on feedback and improve things as we go along.
This is using both user feedback (which comes through lots of different channels, including email, surveys, reviews, our social channels, via account managers and so on!) and usage stats to understand where we can continue improving.
Examples of this include the Impact Score, the algorithm for which we continued to tweak once launched, and upcoming changes to the ‘Artist’ category in our demographics data (many of our users told us it was too broad, so we’re working on refining it).
As you can see, user feedback is incredibly important to us and does have a significant impact on shaping our product and future. That’s why we are always encouraging you to let us know what you think and are incredibly grateful when you do – so please do give us your feedback, opinions, suggestions and requests, we’re listening.
If you ever want to submit a feature request, feel free to let your account manager or the support team know your idea. The more information you give us, the better, and it’s always useful for us to know the original problem (rather than just a proposed solution) so we can explore different ways of solving it.
I can’t promise we’ll do every single one of them, but know that we consider each and every one, from the small tweaks to much bigger feature suggestions.