Designing Effective Communication

My goal in this essay is to encourage technologists and leaders to design communication tools in ways that serve our communities. While the software industry has shown some skill with the individual user experience, we seem rather backwards when it comes to the collective user experience. Designing effective communication requires both!

Attitudes such as “just ship it” and “build only a minimal viable product” pervade modern software culture. In moderation, these make sense. However, it appears many products and services overlook two essential design questions:

1. What problem is your product solving?
2. To what degree does your product solve the problem, under various real-world situations?

Now, when it comes to the problem of effective communication, evaluating your product’s performance “under various real-world situations” is no simple thing. It is complex. To be clear, when I say complex, I don’t mean “impossible to understand”. Complex things are not hopeless — they call for more thinking, not less.

If you are building software to support communication, please design your product to account for human nature. Here are some example questions to guide your design:

• Can your product provide a great user experience for the group sizes and compositions you want to support?
• Can your product maintain the tone and engagement levels you seek?

These examples may or may not apply to your situation. I’d suggest that you make a list of real-world scenarios and how your product performs. This should lead to thinking, experiments, measurement, feedback, and adjustment.

There is no avoiding this complexity. It is essential (due to human nature), not accidental (due to poor design). Put another way, a product designer cannot simply hope that a simple user interface (UI) can prevent the inherent complexities of human nature. For example, both Google and Twitter have relatively simple UI’s, but both know that the nuances of human behavior make their product quite complicated.

I can’t tell you what your product should do or how it should do it, but I can share some lessons I’ve learned about communication in general.

Humans Seek Meaningful Communication

Seeking effective communication is a universal part of the human experience, since it helps connect us and thrive. We seek both information and connection. We seek out environments that promote certain favorable kinds of interactions.

I’ll give some personal examples:

• I avoid loud situations when I need to have nuanced conversations.
• I prefer a group dynamic where all people have room to speak and feel welcome to contribute.
• After some less-than-engaging lectures, I prefer speakers who take the time to get to know their audience.
• After many unstructured, chaotic hack-a-thons, I prefer smartly-organized events like Evan Light’s Ruby DCamp.

These examples are based on in-person experiences, and I don’t think that is a coincidence. More people want these kinds of positive, human experiences to apply to on-line environments as well.

Evaluating Technological Communication Environments

To what degree do technologies provide meaningful communication environments?

Studying public policy in graduate school gave me a new appreciation for collective action and group engagement. When I say effective communication tools, I mean tools that achieve particular communication goals. Goals vary; some include choosing between transparency, fairness, inclusiveness, diversity, efficiency, and more. Communication tools can choose from a wide variety of objectives: conflict resolution, community building, idea generation, planning, consensus-building, or even open-ended time.

Generally speaking, I am inclined to believe that good design increases the likelihood of reaching intended outcomes, whatever they are. In the case of communities and communication, good design is even more important, because that design establishes and signals the initial conditions to its target audience. The right initial design tends to attracts a certain kind of community and encourages norms. Naturally, a community changes and evolves in predictable and unpredictable ways. Effective moderation and coordination often play a large part in the process.

Communication Anti-Patterns

In so many ways, we are just at the beginning of a journey to meld communication and technology. Many of our communication tools (such as Slack, Twitter, or Github) have so much room for improvement. Many fall into common communication traps:

• A tendency towards popularity contests. Instead of maximizing other goals, such as person-to-person connectedness, some tools try to identify the most popular messages. Accentuating majority behavior is not always desirable. For example, many popularity-based systems reward the the most-viewed or most-clicked, which tends to incentivize click-bait. You might consider studying various ways to balance majority and minority interests.
• Defaulting to sequential communication. I’m not saying that presenting information in a linear fashion is necessarily a poor choice, but it is only one design choice out of many possibilities. It certainly should not be the default nor only option. For example, when I think about Twitter, I don’t think of my friends or their opinions as linear — I think of them as an interconnected web. Why must tweets belong in a linear order?
• Drowning people in information. Many communication platforms seem to treat people as infinite consumers of information.
• Ensnaring people. I don’t about you, but my favorite hotel experience is quite the opposite of the Hotel California or The Royale (references to the Eagles and Star Trek: The Next Generation, respectively). When traveling, I often want to enjoy a clean room and to sleep well, and then I want to leave. The same goes for a communication platform. I don’t want to feel like I’m missing out if I turn off notifications for the night. It is possible to build a communications platform so that it is sensible to catch up when I get back.
• Lacking meaningful system feedback mechanisms. I want systems that react, respond, and adjust to user behavior in useful ways. Among other things, I’m looking for granularity. I don’t want to have to unfollow someone completely just because they go on a temporary tangent or rant, but I do want to be spared in the short run. For example, Facebook gives a glimmer of hope in that they allow people to indicate that certain advertisements are not interesting. Platforms should build in feedback mechanisms, both explicit and implicit. Building learning systems is hard, but we’re getting better, thanks to “long-lost” but now rediscovered fields such as reinforcement learning.
• Using ill-conceived and poorly implemented preference aggregration and collective action mechanisms. Talking about voting systems is a guaranteed way to get me riled up. Four years ago, I wrote a short essay titled “Better Online Voting” that still rings true to me, so I would like to share one part of it:

Many … voting systems (even if chosen sensibly) have poor implementations: bone-headed ballots, poor record keeping, poor security, and so on. For example, many sites do not randomize their ballots. Even well-known innovation competitions, such as the Knight Foundation, have given away chunks of money without taking their balloting seriously.

• Seeming ignorance of people’s cognitive biases. People’s actions are imperfect: they are often only short-term approximations that they hope will advance their long-term goals. People take shortcuts, especially when their environment offers them. For example, people tend to read things they agree with. They also tend to avoid or defer difficult decisions and conversations. To equate people’s short-term actions with their long term goals is intellectually dishonest. We must acknowledge people’s humanity and strive to nudge them in directions they will appreciate. This leads to my next point.
• Gathering places set norms, like it or not, and to ignore this is an abdication of responsibility. Many communication platforms are a major part of people’s lives. However, some of these platforms seem unwilling or unable to set norms. You can’t please everyone, but if you refuse to set norms of behavior, you might find that certain parts of your community will dominate. I encourage everyone to watch a wonderful keynote at this year’s Strange Loop: Humanities x Technology by Ashley Nelson-Horstein.
• Unwillingness to better the human condition. How does a platform serve its community? What do people really want and need? How can organizations find a funding (e.g. business) model that aligns reasonably well to the real value-add of a communications tool?

Perhaps the creators of these tools are well aware of how much more is possible. Perhaps they were not designed to achieve the goals I seek. Perhaps the communities they serve don’t yet demand better tools.

I recognize that I am sometimes a harsh and impatient critic of technological and societal change. I understand that building a technology tool that gains adoption and functions reliably is a significant achievement. I also appreciate that building a viable business requires compromises, but I have an intuition that business models do exist to support more effective communication tools. So, while I admit these realities, I think we can aim higher.

System Thinking Applied to Communication

To be clear, when I say communication, I mean a much richer sense than sending messages from Person A to Person B. That narrow sense is probably better described as (merely) transport or conveyance. To achieve a fuller sense of communication, one has to evaluate the system behavior.

In my experience, the software industry often falls short when it comes to thinking about system behavior. There are segments where we do; I will mention two examples. First, in system performance, many of us understand that finding and reducing bottlenecks is more effective than blindly optimizing particular components. Second, in system testing we are starting to understand that we must define meaningful performance metrics based on statistical distributions instead of relying on point estimates that mislead.

What would it mean to think about communication from a system perspective? I like to start with the stakeholders. I don’t claim to know what communication should look like in general. Communities, teams, and people have heterogenous values and norms.

Sub-Goals of Communication

What do organizers and participants (i.e. their constituencies) want when they say “effective communication”? Naturally, various communities seek diverse communication goals. At some point, I would like to attempt to offer a cogent summary of such goals.

I’ll start by listing some common “sub-goals” of communication. I think of these as “rules of engagement”; they don’t necessarily describe where a communication tool wants to lead its people, but they do indicate how the interactions are structured. I list them as pairs to illustrate some common tradeoffs:

• attracting people around a particular topic versus a diverse mix
• protecting anonymity versus ensuring authenticity (e.g. by linking online personas to real-world ones)
• attracting experts and encouraging them to speak more often than novices versus encouraging of turn-taking among all participants
• valuing authoritative information over speculation versus promoting untested, unproven, experimental or fringe ideas

Ultimate Goals of Communication

Next, I want to share some of the teleological (or “ultimate”) goals of communication tools. (The Greek word teleology signifies the end or purpose of something. The New Oxford American Dictionary provides three definitions for the adjective form of ultimate. The first is: “being or happening at the end of a process; final.”)

Here is a partial list of what communication tools seek to accomplish:

• to change opinions or influence a group to some course of action
• to rally like-minded people (e.g. activate a political base)
• to focus discussion on a topic (perhaps by improving the signal-to-noise ratio)
• to forge durable personal connections that transfer to offline experiences
• to collaborate to build a complex product (e.g. with software)
• to build a shared understanding (e.g. sense-making)
• to spread information (usually with some implicit standard of quality or provenance)
• to structure discussion in a logical manner
• to harvest a community’s knowledge into a form that can be consumed by another (Thanks to Lucas Cioffi for suggesting this goal.)

Can you think of examples of which commercially-available tools map to this list? Would you care to add more to it?

Which environments have goals that serve you?

Being aware and critical of the communication tools we use is important. Don’t let any one existing product define your thinking. Reframe it for yourself. Seek out environments that serve your values.

And, lest you think I’m too idealistic, I recognize that there are many less “noble” (i.e. more self-interested) reasons for communication:

• to “capture users” primarily to support a business model (e.g. selling advertisements by aggregating data about your audience)
• to promote a product or service (or yourself)
• to signal status or importance

These are common and very human goals, of course, but I would encourage participants in such communication environments to (a) recognize the true goals of the system and (b) seek out more useful environments where possible.

I will attempt this definition of an effective communication system. It provides: (1) satisfying interactions, (2) using a well-functioning, trusted system, (3) to structure and focus the experience.

How do you (or will you) design communication tools in ways that serve your communities?

If you get nothing else from this essay, I would like you to think about designing a communication tool that serves your community, however you define it.

I would like to gather (or join with) people who want to discuss these topics and improve our technology ecosystem for communication.