Test case: @Mention


As mention’s new Community Manager, I thought I’d take a moment to do the first in a series of case studies on how mention works for us, how it can work for you – no matter the size of your brand, portfolio, or social reach – and how it’s one of the most powerful tools out there for cutting through the noise.

And what better way to do it than by having a look at our own alert for the insanely broad keyword “Mention,” one of the most repeated words on the social web, if not officially the hardest brand name to track online…ever.

At Mention, we’re pretty fearless

Why else would we have chosen the Twitter handle @mention?


Because we have a huge amount of trust in what we’ve built. On an average day, we get thousands of @mentions from users who don’t realize they’re actually, well, @mentioning mention.

@mentioning mention

While this can lead to some pretty funny discoveries, it can get old fast. Especially when you’re buried under a heap of messages like this one:

Screen Shot 2013-10-08 at 7.23.01 PM

Anti-noise, you said anti-noise?

This is where our anti-noise technology comes in. “Noise” is “a mostly unwanted random addition to a signal,” so when it comes to media monitoring, it’s all those mentions you’d rather not be aware of. The ones that take you away from what you should be doing: connecting with your customers, joining the right conversations, tracking your brand.

This noise, especially if you or your company have a more common name, can get loud, preventing you from finding the real signal, the mentions that matter to you. To deliver you only the most important references that make a difference, our technology automatically detects what can be associated with spam, things like automatic tweets, to remove them from the feed. Then, it easily enables you to manually mark mentions as Spam, as well.

Screen Shot 2013-10-09 at 10.05.44 AM

Our smart learning algorithm will then do the rest by putting similar mentions in the spam feed as well, much like how Gmail does it. You can always remove mentions from spam too, especially if you realize they’re actually relevant, and our anti-noise technology will learn to find similar ones again.

Alert settings can always be improved

Most of the time, setting up a simple alert with only one or two keywords is a great start. But for more complex searches, some fine tuning may be needed. You’ll find this under the Advanced settings section. Here’s how our alert is set up, for example: 

Screen Shot 2013-10-08 at 7.07.17 PM

Not only have we gotten incredibly precise with our keywords, but we also often add exceptions that’re not to be taken into account in the search:

Screen Shot 2013-10-08 at 7.26.57 PM

We also block specific sites that we know we don’t need:

Screen Shot 2013-10-08 at 7.07.48 PM

And voila, here’s what our feed looks like without all the noise:


Pure social media WINNING. No more spam accounts, no more Tweet Attacks Pro 2. Just you, your brand, and your customers.

Questions? Comments?

Get in touch anytime on Twitter @xoalexo

Matthieu is CEO of Mention, where he moves all Trello cards to the right and closes deals. He splits his time between Paris and Brussels. Say hi @mvaxelaire.

Subscribe to Marketing + Monitoring Weekly

Get hot blog posts, insanely useful resources, and funny gifs every Friday.