Technology Can Help Us Collaborate

In my previous post, I argued that collaboration isn’t purely — or even primarily — a technical challenge. But technology can help. In this post, I’ll talk about how.

Economics is at the heart of the challenges facing effective collaboration. Technology can help us allocate the scarce economic goods at the heart of collaboration: attention and expertise.


Time is money, or so we’ve been told since childhood. Yet, as often as we’re asked to “pay” attention, we rarely treat our attention like money.

Laziness, irrationality, and social norms prevent us from managing our attention optimally. Instead, we rely on heuristics like first-in-first-out (process a task queue in the order received) and last-in-first-out (focus on what you were just asked to do). Given that time is money — in fact, our time is the scarcer of the two — we can and should do better.

We already use technology to protect our attention. Gmail users appreciate that Google filters out 99.9% of the spam that would otherwise waste our time. Facebook and LinkedIn employ teams of data scientists to detect abusive users and prevent them from hijacking our attention.

Smarter Notifications

But spam filtering is only the beginning. Our communication and collaboration platforms can model the cost and benefit of interrupting us and adapt their behavior accordingly.

We’ve started to see this with products like Google’s Priority Inbox, which uses machine learning to determine which incoming messages are important. It’s not perfect, but it’s a compelling step forward.

The bigger win will come from generalizing this concept across applications and platforms — especially to the many communication and collaboration platforms that clamor for our attention. I expect that we’ll eventually see smart notifications as an operating system feature. They could even be a compelling differentiator for whoever implements them better.

Attention Bond Systems

Another collaboration issue we need to address is the potential disconnect between participants. I may feel that the benefits of your giving me your attention exceed the costs, but you may feel differently. How does a collaboration platform overcome this disconnect?

The most promising technical solution I’ve seen is an attention bond system. Proposed to reduce spam, an attention bond system requires a sender to put money in escrow to ensure that the message is not spam. If the recipient decides that the message is spam, the sender loses the money — either to the recipient or a third party.

While attention bond systems were designed to stop spam, the concept applies wherever there’s a potential disconnect between how the sender and the recipient value the latter’s attention.

Attention bond systems have never seen mainstream adoption — not a surprise, given the infrastructure and logistics challenges. But the economic theory is sound, and I hope collaboration platforms explore ways to apply it.


Much of the potential value of collaboration comes from distributing expertise to where it can add the most value. But how do we identify experts and deploy their expertise efficiently?


The first step is identifying topics and associating people with them. An instructive example is the system LinkedIn developed to discover and tag its members’ skills.

Most inputs for skill extraction are less structured than LinkedIn profiles. But there’s a rich body of work on terminology extraction and topic modeling. Both approaches identify salient topics in a corpus — such as an document repository or messaging system archive — and tag documents accordingly. The document tagging doesn’t need to be perfect, since it will be aggregated to associate people with topics.

But associating people with topics isn’t enough. How do we identify the experts? There are expertise retrieval approaches designed to perform this task based on document analysis, e.g., using a library of research papers.


A more practical approach for non-academic settings is to embrace reputation as an economic good that propagates across a social graph.

The best examples of this approach are Quora and Stack Overflow. Both use voting to determine the best answers for questions, as well as to establish topical expertise in general. The success of these sites demonstrates how a well-designed reputation system can motivate effective knowledge sharing on a massive scale.

Replicating this dynamic inside a modestly-sized enterprise is harder than doing so on a massive public platform, since success requires a higher participation rate. Enterprises can compensate for their size by offering concrete incentives to demonstrate and contribute expertise. The incentives don’t have to be monetary: experts can earn recognition, and their documented expertise can be a factor in their performance reviews.

It takes subtlety to create incentives that are sufficient to motivate participation but not so high that employees try to game the system. But platforms like Quora and Stack Overflow demonstrate that it’s possible.


Attention and expertise are scarce economic goods, and technology can help optimize their allocation. Technology, aligned with the right economic and psychological framework, can improve the way we collaborate.

High-Class Consultant.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store