Unfortunately, you're in a relationship business

Philip Morgan

This is really interesting, and although I'm linking you to what appears to be the full academic paper, you can read the whole thing in probably 3 to 5 minutes: rsa.tandfonline.com/doi/full/10.1080/21681376.2019.1588155

I'll summarize the key points if you'd rather not read the paper right now (even if you don't read the paper, please look at the data visualization that's embedded in it):

  1. Stack Overflow (SO) makes its entire content archive available here: archive.org/details/stackexchange This presents an interesting opportunity to study and visualize the patterns of communication between SO users. The paper's authors take advantage of this opportunity.

  2. They ask the question: how centralized vs. distributed is collaborative knowledge creation? They choose to think of multiple contributions to a single SO question (either on the question or answer side) as a collaboration, and they draw lines on a world map between the geographical location (if available) of SO users who collaborate.

  3. The paper's authors use the resulting map, which compares SO activity in the year 2009 to the year 2017 using a random census of 1% of all activity, to advance a hypothesis that the Internet has increased rather than decreased centralization, at least to the extent that Stack Overflow can reveal these patterns.

The paper's authors definition of "collaboration" is OK but also debatable. They define collaboration thusly: "we constructed a network: similar to scientific collaborations, two cities are connected if users from the cities jointly contribute to a question, either in posing the question or in providing an answer." That's not how I think of collaboration, but also, it's a useful way of viewing patterns of sharing knowledge.

If your worldview of the Internet is similar to mine, then this research presents a piece of evidence that contradicts your worldview, which can feel upsetting. Or alternately, this new evidence provides an opportunity to increase the fidelity of your worldview, which is a good thing.

Either way, it's somewhat of a surprise to see the Internet -- something you think of as a democratizing force -- host more centralization. It's surprising to see something -- again I'm talking about the Internet here -- you view as "wanting" to increase inclusiveness actually generate a less inclusive conversation.

Is this because of toxic, Silicon valley "bro culture"? I and this email list aren't here to dig into those issues. Rather, this is my opportunity to remind you that, as I often like to say: unfortunately, you're in a relationship business.

I say this with a grin because many of us hate this reality, or resist it in some way. I've lost count of the number of times I've politely interrupted a client who is 5 minutes into detailing how they plan to use an email to respond to a complex, difficult situation with their client and said, "Just call them. I can help you prepare for the call, but please just call them. This will turn into a dumpster fire email thread and a mass grave of misunderstood intentions if you don't." And I've lost count of how many times I've been right to give this advice.

Unfortunately, you're in a relationship business. :->

Relationships vs. expertise

There's an interesting tension between relationships and expertise.

The platonic ideal of the expert is1:

  • Their expertise has value.

  • This value increases the demand for their time and access to their expertise.

  • This increased demand causes them to put in place various filters: high price for access to the expertise, gatekeepers for access to the person and their time, habits and norms that maintain or improve the expertise asset at the expense of availability to the public or other assets.

Experts tend to be inaccessible, except if you pay a high price for access or you make it through a dense layer of intentional filters.

On the other hand, building relationships often requires a generous investment of time. The trust that enables relationships takes time to build. Furthermore, someone usually has to "go first", meaning one side of the relationship has to take a risk by trusting the other side more than the relationship would seem to merit. Relationships grow with investments of time and risk.

If you're an expert conforming to the platonic ideal of an expert, how do you build relationships?

I'll be brief here, since this is not the main point of this article:

  • A relationship with an expert is not a peer relationship. The expert has a form of power you do not, but you want access to. This power reduces the time it takes to build trust in proportion to the magnitude of the power differential. I can mow my own lawn, so the lawn service has relatively little power in the relationship with me, and so it takes a relatively long time for me to come to trust them. I cannot diagnose some major illness I might be suffering, so the physician has relatively a lot of power in the relationship, and so trust (in a limited sense) can be built much more rapidly between my and the physician.

    • Consulting takeway: Find a way to deliver a quick, impressive expertise-based win early in your relationship with clients or qualified prospects. A common way to do this is to throw the previous consultant under the bus by criticizing some obvious flaw in their work. Unsolicited teardowns have the same flaw. I fucking hate these approaches to trust-building because Consultant B doesn't know what constraints Consultant A was operating under, so it's almost always a cheap shot at the expense of a fellow professional. Please find a different way to deliver a quick, impressive expertise-based win.
  • Experts have more degrees of freedom in their personal life that allow them to invest in relationships in powerful ways. For example, a successful expert can take a risk by trusting the other side more than the relationship woul d seem to merit by saying, "Hey, why don't you spend a week at our vacation cabin in the mountains some time this summer?" This is an example of a relationship-building mode that might be more available to experts than other folks.

    • Consulting takeaway: What ways of rapidly building trust with clients or qualified prospects might your expert position afford you? This is less about using the expertise itself and more about using the by-products of that expertise (money, discretionary time) to be generous in interesting ways. Ex: This hasn't happened yet, but now that my daily emails are switching to a paid subscription, I plan to compile 3 to 4 topical anthologies every 2 months and send them to qualified prospects and clients. These anthologies would be made up of the best few paid emails from the last 2 months on the topics of specialization, lead generation, and cultivating IP, made into 1 e-book per topic, and distributed to prospects/clients based on what specific challenge they happen to be working on. This is an example (hopefully!) of building trust using the discretionary time that I have because of my expert position.

Experts can build relationships, of course, but how they do it looks different because of the power differential, and because they have different degrees of freedom for investing in relationships.

What the SO map tells us about relationships

Let's weave these threads together.

One take on the Internet is: most of the world's population is connected and accessible in some way, either directly through messaging (email, etc.) or indirectly through an aggregated platform (social media, etc.). For quite a while, this has been my operating worldview. It's very much a techno-utopian worldview, and certainly a simplified way of understanding the world, as all worldviews necessarily are.

The paper I linked to at the top of this article puts a significant kink in my operating worldview. Over time, the actual patterns of connection between SO users are looking less like my worldview, not more. One of the following things might be happening:

  • Yes, the Internet connects in a universal, democratizing way, but few people make use of this capability. In other words, perhaps we just prefer to not connect in a universal way, and instead prefer to connect with others like us.

  • The Internet does not actually connect the world in a universal, democratizing way. Access is more centralized in reality than it is in my techno-utopian worldview. I have evidence that this is true from my 2 years at the Oregon coast from 2011 to 2013, where the best Internet connection available to me was 1.5 Mbit DSL or somewhat faster satellite, both completely unsuitable for even a low-quality video call.

  • Reality is more complex, and cannot be explained by simple single-factor explanations.

If you think of that data visualization from the above-linked paper as a map of relationships rather than as a map of the Internet's democratizing capacity, it makes a lot more sense.

Said differently, the map seems to contradict view #1 and support view #2 below:

  1. View #1: The internet broadens access to information and connection between people.

  2. View #2: Unfortunately, you're in a relationship business, and this is visible even when you look at SO activity.

I don't want to dilute my point too much, but I'm compelled to point out that single-variable answers/solutions are rarely effective or complete answers/solutions. People are migrating to cities, and the tipping point where more than 50% of us will live in cities is not far in the future at all. So maybe this migration to cities does more to explain things than anything else?

The world is a big complex place. Resist oversimplified answers and solutions!

And, invest in relationships. Because after all, you're in a relationship business whether you like it or not.


  1. Please read David C. Baker's "The Business of Expertise" for more on this.