How a social network worksThere is a fundamental difference between a search engine and a social engine. It can be summed up in two simple statements:-
I know what I want, go and find it for me
—-the search engine user
I have no idea what I want, show me something interesting
—-the social network user
It sounds incredibly simple but the reality is that a search engine and a social network really work on two very different principles. The search engine gets to start with a keyword while the social network, it gets nothing.
So how does a social network figure out how to show users stuff that they might be interested in?
Well one of the things it does is it triangulates
It’s all about who you know
Business is about who you know, not what you do. My uncle used to say that to me when I was young as if it was some sage like advice passed on through generations. Turns out when it comes to Digital Marketing he was right.
When a search engine does it’s search it has a keyword to start with. A social network gets no direct input from a user but it does have information upon which it can base an algorithm. It knows who you are. It also knows who you know.
Know the person
Firstly a social network has to know who a user is. Pretty much all networks will try to identify users through their email address and phone number. They also collect information about what they look at, what they click and how they behave in order to build up a picture of them. This is common knowledge and most regular social media users understand that if they look at a lot of one type of content the social network will try to show them more of that type of content.
A social network learns what a user likes and what they are interested in.
Triangulate their connections
The other thing that pretty much every social network asks a user to do is connect to other users in some capacity. Whether they are friending, following, connecting or even partnering, the social network will keep track of these connections and compare the information it stores about the user with the information it stores about their connections.
A social network learns about a user’s personal circle of connections and then it keeps them in a big list.
Now everyone has guessed that a social algorithm bases its results on prior interest. What is not so well known is that a social network also looks heavily into a user’s list of connections.
Same people, similar interests
One of the assumptions that a social network makes is that a user is similar to the people they connect with. For the most part that is actually a reasonably accurate statement.
Sure, your granny may not be interested in the latest death metal band you are into, nor are you particularly receptive to the knitting patterns that have started to appear in your feed but for the most part, people tend to hang around in groups with similar interests. It’s just human nature.
A social network exploits this concept and uses it to target and filter content.
If a users connections are interested in this thing then there is a reasonably strong chance that they are as well. The more of the users connections that are interested then the stronger the correlation towards personal interest and the more likely the chance that the user will interact with the content.
This is what a social network wants to happen; it wants the user to tell it what they like. If the user won’t tell the network then it will just find out from their friends.
Who a user connects with will determine what they see. What a users connections do to each other will also influence their feed.
So what’s all this Triangulation then?
Let’s take an example.
Four people are all friends on Facebook. One of those people decides to post an update about something that interests them.
Two of the other three friends see this on Facebook and leave a comment and a like on the update.
They interact with the content.
The social network, in this case Facebook, detects this interaction and triangulates their friend lists. It looks down each individual list of friends that belongs to each of the three users who interacted and then tries to find other friends who exist on each of the lists.
It triangulates their friends list to find common shared friends.
In our example there is a fourth person. This person never saw the content and never interacted with the update. They are, however, connected to three people that did.
Since the fourth person has three friends who all interacted on a single item of content the social network will almost certainly notify them in some capacity that this happened. It might be that the update is now more likely to appear in their feed. If enough friends triangulate then it will include the update in a regular email.
When you get an email that says “look at what your friends are interested in” it’s really not kidding!
How does Triangulation affect reach?
There is a general misconception that on social media the numbers are all powerful. Getting a Retweet from a profile with 100,000 followers is far more valuable than getting a Retweet from a profile with 1,000.
The profile with the large number has greater natural reach but if it is an account that has absolutely zero shared followers with the person doing the Tweet then this is actually inferior to the second profile. If the thousand followers all have an established relationship with the content author then even though the numbers are less the strength of that link will dominate.
100,000 followers may, if we are lucky, get something in their timeline due to a Retweet. It will not be prioritised and the chances of the follower spotting it are low. The 1,000 followers on the second account who all follow the content author will definitely get a timeline mention. If a few more accounts interact with each other then they will get a highly visible prod from Twitter in the form of a sidebar mention or a “what’s happening” email.
Quality over quantity
This kind of interaction driven reach is far more conducive to trending. As more common shared accounts get involved the algorithm is far more likely to be able to identify and prioritise its reach to targets that are actually interested in that subject.
If a lot of quite random and disconnected accounts interact the algorithm has very little overlap to work with. Content ends up being shown to the wrong people and these people who are not in any way likely to interact with it. This kills the trending process stone dead.
So what is the best strategy?
Please understand that this is not an exact science and there are many other factors that will affect reach, interaction and general success metrics. The quality of the content being posted being the main one.
What comes out of the understanding of social media triangulation is that the relatedness of those who interact with content matters. Numbers still count but if those numbers also have a tightly coupled relationship between each other then this dramatically increases the effect of the interaction.
The algorithm starts to work properly and content gets to the right place. More people interact and that continues the process.
What needs to be done?
Digital marketers need to do two things:-
- Firstly they need to promote interaction, even if it’s a bit random. There is still a high likely hood of random triangulation on content that has naturally achieved a lot of interaction.
- Secondly marketers need to concentrate on tightly coupled groups that have a high level of connectedness. This is one of the many reasons why getting an influencer (preferably related influencers) to interact is such an effective strategy.
By understanding that not all interaction has an equal value it is possible to hit groups with highly targeted content. It may never get a massive amount of interaction but if a few of the group do then the rest of the group will get a notification. If the content is lined up and on message then it create a small niche trend. This is something that is not global but that successfully ripples through a community.
Understand that triangulation is important. Fostering controlled interaction between highly coupled groups of people is what Yatter is all about.
One of the things that social media is very effective at is the specific targeting of individual sales targets in a business to business scenario. Whether it’s the purchasing director, managing director or just a massive big whale, the digital marketer desperately need to speak to that person and get them to sign that million dollar deal.
A very effective technique to get that target to notice you is to get your content in front of their nose repeatedly so that they gain brand recognition of what you are selling. There is nothing better than encountering a sales target at an event and finding out they have already seen and heard of you and what you do.
How do you make sure that your content always turns up in their feed?
Well, get your colleagues to make a genuine attempt to connect or follow the target. Get as many people as you can to establish a relationship to the person you want to syndicate content to. Then write some content you think the target might be interested in and get all those people to interact with it. Go crazy on the comments, likes and shares. Triangulation will kick in and because your target has numerous connections to people in their connection / follow list the content will be prioritised over all others.
The dirty little secret among digital professionals is that they can engineer triangulation. What they also realise is that a good marketer needs to be prepared to put the effort in.
This can be a very effective technique on LinkedIn and Twitter but in all reality it is only worth doing if you have a few very specific targets. Realistically this restricts the technique to business to business scenarios. Often it is done off the back of a few of a sales team meeting the target at an event and following up with a connection. If this a a regular thing for you then make sure your whole team does a connect.
SummaryTriangulation is when a social network scans your friends / follow list and compares them to the lists of other people who have interacted with your content. If someone who follows an author also follows people who interact with the authors content then that content will be prioritised in their feed.
Good digital marketers know how to stimulate triangulation.
Free accounts for a limited time only!
Claim yours today