Is Email the Solution to Information Overload

Information overload prohibits you from "focusing on what matters most." Email is the primary cause of overload because you have too many email message to process. Solve the email problem and you will conquer information overload.

This is what IBM, Microsoft and Google want you to believe. All three companies recently released products to help you cut down on email-related information overload.

Here is how each company touts their new offerings:

IBM Verse offers a faster, better way to manage business communications across devices, organize inbound and outbound information, and focus on what you need most.

Google Inbox is a fresh start that goes beyond email to help you get back to what matters.

Microsoft Clutter is designed to help you focus on the most important messages in your inbox. It uses machine learning to de-clutter your inbox by moving lower priority messages out of your way and into a new Clutter folder. Ultimately, Clutter removes distractions so you can focus on what matters most.

Focus on what matters most …. Is it an incredible coincidence that all three companies identify the same need? Or is this problem so commonplace that the demand for a solution is abundantly obvious?  Will "fixing" email really alleviate the problem of information overload and let you focus on what matters most?

I think not.

My own academic research evaluating the perceived causes of information overload in the pre-email era to the post-email era found that while email has become the poster child for information overload, the real perceived causes are broader. The research, “The Impact of Intra-organizational Email Usage on Worker Information Overload,” found that all forms of contemporary computer-mediated communications contribute to the perception of information overload, and that includes more than just email. 

These technologies include a host of enterprise social networks like Yammer and Jive, unified communications tools like Skype and Lync, and text messaging services. Additional sources of overload include a new generation of operational applications and cloud services like CRM, ERP and project management solutions, as well as document management systems like SharePoint and Dropbox -- all of which generate a steady stream of updates and notifications.

Fixing the email problem will not alleviate information overload and will not allow people to "focus on what matters most."  Fixing email is like plugging the crack in the dam after the water is already over the top. It’s too little and it’s way too late. 

Why is Email the Problem?

Email remains information overload’s "bad boy" because it is the only easy way to share text messages and documents with colleagues, partners and customers.  It’s hard to believe there was a time when email was a walled garden technology relegated to restricted Lotus Notes, Microsoft Exchange, Hotmail, Gmail and Yahoo Mail environments. The widespread adoption of the Simple Mail Transfer Protocol (SMTP) as a standard changed all that. So while email is standardized, many of the other technologies remain proprietary. 

What’s the Solution?

NYU New Media Professor Clay Shirky is famously quoted as saying that we don’t have an information overload problem, but rather a filter failure problem. He correctly states that stemming the tide of information is not realistic, so effective filters are needed to separate out the unimportant stuff. 

Which explains why the big software companies are concentrating on building better filters for email. However, email is only one source of overload; the real question is, “Can we filter information from all sources including email, social networks, unified communications, and applications and cloud services?”

I think the answer is yes, but it will require more than a sophisticated filter. It will require a completely new way of organizing information. 

Taxonomy, Tags and Topics … Oh My

Combining information from multiple sources entails classifying items into meaningful categories.  Two popular ways to do this today involve predefining a set of topics and then tasking individuals to assign items to these categories (taxonomy); alternatively, allowing an individual to assign an item with any label they deem meaningful (tagging). When searching for information, individuals can filter by topics or tags to reveal only those information items that match the filter.

It’s easy to see why neither method scales to solve information overload.

Taxonomy projects suffer from a host of problems, including deciding how many categories to define. Create too many topics and people drown in too many options, define too few and people won’t find information since each category will contain too many items to sift through. Tagging doesn’t scale because different people call the same thing by diverse names; it is difficult to find information employing someone else’s tags. 

Automation to the Rescue

This mandates a new approach of organizing information. The new approach must introduce a consistent, automated way to extract meaningful topics from information items. Consistency eliminates the politics of taxonomy and the variance of tagging.  This approach must fulfill the following requirements:

  • Automatically extract topics from information item across many sources, including social networks, email, unified communications and messaging systems, as well as from operational applications and cloud services. This can be done via machine learning and Natural Language Processing.
  • Match topics across the different systems. This is complicated for the same reason that tagging doesn’t scale, because individuals and systems use diverse naming conventions for things like customer names, contact details and project status.  However, by applying similarity algorithms, computers can make some educated guesses, even if they will make mistakes.
  • Allow individual users to tweak mistakes made by automated processes. No automation engine will always get it right. Human intervention is needed to add corrections.

When this is done, individual users can apply topic filters across all sources of information. Since topics are built from the actual data and matched using similarity algorithms, individuals now have access to the most objective categorization possible.  For the first time, people can truly focus on what matters most. 

Are We There Yet?

We’re not there yet. The IBM, Microsoft, and Google announcements clearly reflect a market still mired in misconception that email is the root cause of the information overload problem.

Initial strides are being made. Microsoft took a first step with Delve by going beyond email and introducing documents into the information mix. Delve also looks at Word, PowerPoint, Excel, PDF documents stored in OneDrive for Business or SharePoint Online when considering filters. 

But this is merely a first step, because a real solution to information overload must incorporate a filtering capability that considers the complete set of information sources.  In the consumer space, Flipboard through its recent acquisition of Zite, has demonstrated impressive results by automating the categorization of items from many news information sources. In the enterprise, some innovative companies are taking on this challenge, so we can expect to see some big strides in the near future.

Shirky is indeed correct.  “It’s not information overload, it’s filter failure.”  But building that filter will require a completely new way of organizing information. Those who figure it out will finally conquer information overload and truly let "people focus on what matters most." The IBM, Microsoft and Google announcements are a good first step, but there is a long way to go.

Creative Commons Creative Commons Attribution 2.0 Generic LicenseTitle image by  staffan.scherz