Although last month’s launch of the new Trackur social media monitoring dashboard was the result of more than 6 months of hard work, we’ve not been resting on our laurels since then.
Nope, in fact, we’ve been hard at work on a brand new feature that we think you’re going to absolutely love. Ready?
Today, we are announcing the addition of automated sentiment analysis to ALL PAID PLANS!
That’s right, starting at just $18 a month, Trackur is bringing fully automated sentiment tagging to the masses.
As you know, I’ve spent many years pushing back on automating sentiment analysis. It’s my philosophy that only a human can accurately navigate the nuances of the human language and understand with 100% accuracy whether a tweet, update, or posts is positive or negative. However, many of our users have asked for a little help with this task, so today we’re delivering.
There were three ways we could have tackled this feature:
- Analyze the sentiment for the keyword in general. An aggregated score of the sentiment for all mentions of your keyword. Useful for getting a quick snapshot, but useless in helping you understand which individual results were positive, negative, or neutral.
- Analyze the sentiment of the document/result and basically determine whether the entire document has a positive or negative theme. More useful, but loses value if you’re tracking a keyword such as iPhone in an article predominately about AT&T. If the article is littered with criticism of AT&T but praises the iPhone, you’d get a false score.
- Analyze the document for the sentiment of the keyword within the document. Perfect! You get a sentiment score that is weighted more towards the keyword that you are actually monitoring.
Can you guess which one we chose to implement? Yeah, that was a loaded question.
Trackur’s automated sentiment analysis looks at the specific keyword you are monitoring and then determines if the sentiment towards that keyword is positive, negative or neutral with the document. That’s weighted the most in our algorithm. We then also take into account the overall sentiment of the document, but to a lesser extent. We believe this gives you a much more accurate picture of the sentiment towards your saved search.
Now, when you view your dashboard, you’ll see the sentiment icons start changing automagically before your eyes. We’ll continually work to improve accuracy, but if you see any item that is mislabeled simply click on the correct sentiment icon to change it.
Automated sentiment analysis rolls out today for all paid plans. It’s our first iteration, but the accuracy is pretty high. We have plans to continually improve it, but we wanted to get this in your hands ASAP. And, the good news is that we’ve been able to add automated sentiment analysis without any price increase.
Side note: We’re looking at ways to include this in our API, and may also offer automated sentiment analysis as a stand alone API in the near future, so stay tuned.
Trackur may be a small team, but we’re big on features. And we just love delighting our customers. Almost 50,000 of you have registered for Trackur and we want to make sure we keep evolving and delighting you with new features.
And, we’re not done yet.