Before Twitter, the hashtag (#) used to be the lonely and desolate button on your phone reserved for ending numeric inputs and employee extensions.

Now the hashtag is a conversation starter, aggregator, and overall good time.  But what can we really gleam from this rags to riches story? Are advanced data-mining tools passing the hashtag by?

Where do you fall on this argument?

Why we may need to look beyond the hashtag:

  • #Noncompliance – not everyone uses them or uses them regularly. Leaving important content and conversations outside the grasp of a simple web scraping tool.
  • #Misspellings – the Achilles heal of the whole system really. If it’s spelled wrong or the wrong hashtag is used, the content again will be missed.
  • #verylongandunreadable hashtags – with only 140 characters to work with, your content real estate is already limited in Twitter. Having a long hashtag hurts one’s ability to comply with using it.

Want an example, hit up the #sourcinginnovationsummit on twitter. #notkidding.

  • #Autocorrect – this is slowly improving as apps are programmed to automatically populate frequently used hashtags. However, it has the potential to exacerbate the misspellings issue, and again, not all apps that interface with Twitter have this functionality.

So the hashtag might be here to stay, but sophistication in unstructured data mining and analysis is allowing us to move “beyond the hashtag”. Because the real world has autocorrect and spelling errors. We needed a better way to mine information, glean insights, and justify assumptions and access relevant information outside of a hashtag.

  • Complex query builders – our tools for mining keywords, strings of keywords, and sentence fragments has greatly improved the ability to pull relevant information outside of the hashtag.
  • Natural language processing (NLP) – is getting better. It is another way to use computers to understand the meaning and context of phrases and words. Note that some of NLP is executed by machine learning below, but now we’re getting into the weeds of things.
  • Machine learning – the folks at Google know something about this. By using computational power and statistics, The ability to have machines read and understand the context of language is going to make huge strides in the next 5-10 years, and the benefits are going to be insightful information pulled from text and messages.

While technology is allowing us to look beyond the hashtag, you should use a consistent, dedicated hashtag methodology for your event. This will ensure that you and your attendees have a dedicated discussion to contribute too.