LinkedIn and Narrow Channels
When you add random people on LinkedIn, it is mildly advantageous for you, but a massive disadvantage to your other connections
About a year ago, when I was still in a comfortable corporate job, I decided to “clean up” my LinkedIn. I sat diligently for 1-2 hours on a weekend, going through my entire contact list, and removing anyone who I didn’t personally know. This included headhunters, people I might have weirdly bumped into some event once, people whose posts landed up in my feed and which I had ended commenting on, etc.
I don’t remember the numbers, but I think I had brought down my connection list from ~1800 to ~1400. I remember feeling extremely satisfied at having done that.
Six months later, I was regretting it. I wanted to look through my LinkedIn list, to see who I could chat with for “research” into my business. Suddenly, a “clean” contacts list was a bug rather than a feature. I proceeded to start adding people back. And possibly having forgotten about it, some people I had been connected to earlier didn’t accept.
Now, four months into official entrepreneur life, my LinkedIn contact list has become “dirty” again. And this is a feature rather than a bug. It had started with venture capital associates, who had connected with me well before I was ready to start raising funds. As it happened, I’ve met most of them 1-2 months after I had already been connected to them on LinkedIn.
Then, I’ve been putting out “content” on LinkedIn (though most of it are links to this blog, which I think LinkedIn shadow-bans). As a result of this, I’m getting more connection requests. There are possibly prospective employees; possibly prospective vendors; possibly prospective customers. And so forth.
Bandwidth
Earlier this month, we were preparing our list on venture capitalists to reach out to. The major ones in India were not a problem - between Manu and me, everyone is a maximum of two “strong” degrees away. We are also hitting some US-based VCs (we are headquartered there, and plan to sell there, so makes sense to have investors from there), and this is where things started getting “interesting”.
There was one VC I wanted to connect to, and who was two degrees away. There were 10 mutual connects. I quickly guessed which of them is most likely to know this VC (one startup founder in the US who recently raised a round), and asked him for the connect. It is another matter that this VC didn’t want to talk to us (they have invested in companies broadly similar to ours, and want to avoid any conflicts of interest), but that there were so few common connects made it really easy to reach out.
Similarly there was another VC I wanted to connect to. This time there were 70 mutual connects. And none of these was an “obvious introducer” (part of a portfolio company), and I had to go through a few cycles of “can you introduce me to this person?” “no, I don’t know him well” before I actually got the intro.
In other words, where the channel was “narrow”, it was far easier to get the intro than when the channel was broad. There was no decision fatigue. I didn’t mind so much if the potential connection was “weak”. I went ahead anyway. The broad channel was more tricky!
Maybe this is why venture capitalists insist on us going after an extremely narrow target market to begin with - with a narrow channel, we will dig deeper (ok that’s a bad mix up of metaphors), and have a greater chance of doing something.
Contamination
I realise the negative externality of adding random people on LinkedIn is not going to be borne by me (for me they are all “option value”), but by my connections - because you add “weak ties” on LinkedIn, people’s channels get unnecessarily broader, which leads to potentially weaker connections.
For example, someone might want an intro to someone whose random request I’ve accepted in the last two months - I clearly don’t know these people well enough to make a strong introduction, and so I’ll refuse to do that. And this will imply more work for the person who wants the intro. They unnecessarily got their hopes up thinking I will be able to introduce them!
Strength of connections
As a corollary, I’m thinking - if I want to connect to a person X, and there are 3 potential connectors A, B and C. For the intro, I should choose the person who (potentially) knows X the best. Even if my connection to the introducer is relatively weak, as long as the connection between the introducer and the introduced is strong, I have a good chance of a conversation.
If the introducer doesn’t know X well, it doesn’t matter how well I know the introducer - the intro will be weak (or - the “lead is not going to be strong”), and I will not get the outcome that I’m looking for.
(Of course, the ideal case is that someone I know very well also knows X very well. The probability of that, though, is low, and thus one needs to plan for contingencies.)
In that sense, if one of these “random people” whose requests I’ve accepted asks me for an intro, I’d still be inclined to help them. But I may not help other people to connect to these people.
In other words, “weak ties” might result in a directed graph. That is a discussion for another day.