LinkedIn data analysis

Do you think you are getting too many LinkedIn requests? I had a sneaky suspicion that I was getting a bunch of requests, but I wanted to empirically check how many I was receiving.

So I tracked every request I got for a month.

In the last month I received 25 requests – which is almost once per day (and more than once every business day). A summary of all the requests I received are below.

Some highlights:
# Requests: 25
Number where I was the intended target: 2
Number that I initiated: 0
Number that I was just a link of the chain: 23

Of the 23 …
Number that I forwarded: 9
Number that I declined (thought they were inappropriate): 14

Of the 25 … number of hops:
Two: 2
Three: 2
Four: 6
Five: 15
(commentary: because I was hop #3 in a 5 hop chain of most requests, I was less likely to forward the request because I have no relationship either to the person initiating the request or to the intended target)

The data:

2/15/05 — 5 hops. i was hop #3. someone wanted to recruit a former CEO for a manager position in Teaneck NJ. i declined.

2/13/04 — 3 hops. i was hop #2. one VC wants to meet another VC. i forwarded.

2/10/04 — 4 hops. i was hop #2. good recruiter looking for a candidate. i forwarded.

2/9/04 — 5 hops. i was hop #4. someone wanted to pitch work to someone across the country. i declined.

2/9/04 — 4 hops. i was hop #3. someone wanted a job. i declined.

2/8/04 — 5 hops. i was hop #3. someone trying to pitch Microsoft. i declined.

2/6/04 — 4 hops. i was hop #3. some guy wants to connect to a former coworker. i forwarded the request.

2/6/04 — 4 hops. i was hop #2. friend wants to sell something to Washington Mutual. i forwarded the request.

2/6/04 — 5 hops. i was hop #3. someone trying to reach the asst to the CEO of Starwood to sell them something. i declined.

2/5/04 — 5 hops. i was hop #3. someone trying to reach the CEO of Princeton Review to pitch them something. i forwarded the request.

2/4/04 — 3 hops. i was hop #2. person trying to connect with an old business colleague. i forwarded the request.

2/4/04 — 5 hops. i was hop #3. someone i do not know wants a job from someone i do not know. i declined.

2/4/04 — 5 hops. i was hop #3. someone wants to learn about a company five hops down the chain. i declined.

2/3/04 — 5 hops. i was hop #3. Some guy wants to sell camera phones into starbucks. i declined.

2/2/04 — 5 hops — i was hop #3. some guy wants to introduce a new piece of software to another guy. i declined.

1/29/04 — 2 hops — i was hop #2. someone i never heard of is moving to SF and wants to connect. i declined.

1/26/04 — 5 total hops. i was hop #3. guy looking for a job. i declined.

1/26/04 — 5 total hops. i was hop #4. someone looking to get into the wine and food industry. i forwarded the request.

1/23/04 — 5 total hops. i was hop #4. someone looking for consulting work. i declined.

1/22/04 — 4 total hops. I was hop #3. looking for a candidate to hire. i forwarded the request to a friend who is the potential candidate.

1/21/04 — 5 total hops. I was hop #5. Guy wanted to reach me. wanted to get involved in Lead21. i accepted him. he immediately became a Lead21 member

1/19/04 — 2 hops (guy requested me directly). wanted to sell me something. i declined.

1/19/04 — 4 total hops. I was hop#3. some random guy wants to meet michael yang. i declined.

1/17/04 — 5 total hops. i was hop #3. someone looking for a job. looked unqualified. i declined.

1/14/04 — 5 total hops. i was hop #4. job seeker wanting to get in touch with a recruiter friend of mine. looked legit so i forwarded it on.

6 thoughts on “LinkedIn data analysis

  1. Cynthia Typaldos

    For most social networking, once you are more than 1 degree away from a target, the whole thing gets pretty silly. That is, it makes sense for a colleague to introduce two people that s/he knows (you are A, your colleague is B, B knows colleague C and introduces A to C ), but little sense for A be introduced by C to colleague D (where D is a colleague of C but not of B or A).
    There are 3 types of items that can be passed up and down a social network: physical items, information, and introductions. A three dimensional matrix can be built of these three items vs. node motivation & need for closure of feedback loop. If the node is not motivated, nothing will happen despite a million links between that node and the other nodes. If a feedback loop is needed it doesn’t matter how many nodes the item successfully passes thru but only whether the first node, who initiated the activity, ever gets a response.
    Physical item – example is a virus of an infectious disease. Nodes are all powerfully motivated by natural selection (that is, the virus is designed to infect others), but no feedback loop is necessary.
    Information item – example is salesperson wants contact info for target customer. Nodes need to be given a reason for motivation (reputation enhancement could be a motivator), and it is essential that once the node with the information is discovered, the feedback to the initiating node is completed.
    Introductions – example is a job reference. Introducing node must be 0 degrees away from one introducee, and 0 or 1 degree away from the other introducee.
    Continuation is here:

  2. Scott Allen

    Actually, LinkedIn’s data suggest otherwise. I can’t disclose the exact numbers, but I can say that the acceptance rate by the target of introductions that make it through 4 degrees of separation, is well over than 50% — much higher than I expected. Now, it may be a much smaller percentage of the the total requests initiated, but that means that the network is working like it’s supposed to, as a collaborative filter.
    Trust is more transferable than you give it credit for. Take for example the case of Reginald Warlop, who got his current job as a VC at Ariadne Capital in England (after moving from San Francisco) through four degrees of separation: a CEO, a board member, and an EVP, on the way to the final contact.
    How well it works depends on how strong the trust is with the relationships you list. Everyone on my list is someone I would feel 100% comfortable referring people from or to. That doesn’t necessarily mean that I feel that way about their contacts — that varies from person to person.
    The trick is to teach people to be sensible about the kind of requests they make. For example, to ask Auren to “just connect because they were moving to San Francisco”, with no more basis than that, is noise.
    But good requests CAN travel effectively, and be filtered effectively, in a network of moderately high-trust relationships.

  3. David Nordell

    My own experience with LinkedIn, over about three months of active use since joining, is that I’m getting about one request every two to three days, mainly from someone I know directly or from a direct contact of one of my contacts (in other words, I’m hop 2). I’ve rejected a couple on the grounds that the person I was asked to refer to was unsuitable, and made suggestions to the originator about more suitable recipients. I’ve rejected several from a friend of mine who’s job-hunting and told her to rewrite them, because the style of the request was simply unprofessional: I also felt it would reflect badly on me to pass on requests written like that. I just rejected one from another friend yesterday and told him to rewrite it because of a stupid typo (and also a lack of information) that reflected badly on him.
    Similarly, I’ve received a number of requests to me that were unsuitable: I’ve replied politely to a couple of them, and even taken time for telephone conversations with the people to give what advice I could; and I’ve replied ‘not interested’ on other occasions.
    The other thing I’ve done where possible is to suggest connections to some of my own contacts (it’s also happened to me that contacts of mine suggested sending requests to other contacts of theirs).
    What conclusion would I draw? I don’t think the statistical analysis is worth very much. What matters is to treat LinkedIn similarly to any other form of social or professional networking: to be honest and helpful with people’s requests, to take the initiative in helping the people you know, and also to see passing on well-founded requests as a service that you’re giving to other people, whether because you might get something back in the future or purely out of altruism. Actually, I don’t think altruism is really the point: I’m one of the best connected people in Israel on LinkedIn (and I’m also active in several other networking services), which raises my professional profile and creates the perception that I’m worth knowing (which is of course true).

  4. Konstantin Guericke

    The way we designed LinkedIn, the user experience directly mirrors the quality of your connections.
    It is key to be connected to the right people: people who can help you meet someone through a strong recommendation when you contact someone through them, but also people for whom you are willing to do the same–take time to help them get in touch with someone.
    If you are connected to people you don’t know well, then forwarding becomes a chore. If you choose them well, it’s a joy because it’s great to help someone you know and trust. And it builds relationship equity. Using LinkedIn, I now facilitate at least ten times the intros I used to do in a fraction of the time it used to take because LinkedIn makes the requester do 90% of the work.
    And there are never any 5-hop requests–the maximum is four degrees, which means a max of 4 hops and 3 connectors.
    ONLY if you are in a 4-degree request AND you are the middle connector, do you know neither the sender nor the receiver directly. Given that a lot of requests are for 2 or 3 degrees and that there are always two connectors who know the sender or receiver, this is not a common stage. However, it’s a very important one since BOTH the sender and receiver are friends-of-friends (actually, often I have already heard about either the sender or the receiver through past conversation with my direct contacts, so they are not total srangers). And a connection on LinkedIn means that you not only trust a person, but also their judgement in the company they keep.
    So, next time someone asks you to connect, my advice would be to check:
    1) Are you willing to take a 5-10 minutes per year to help them with intros?
    2) Would you enjoy helping them because you think they’re great?
    3) Have the quality of people they’ve introduced to you generally been good, so you can feel good about forwarding requests for their direct contacts?
    4) Do you know what they want and need well enough to know which requests to forward to them and which ones to screen out?
    5) How much can this person help you with achieving your career and business objectives?
    Also, if you have forwarded requests, be sure to check the results under the request tab. There is a sub-section called “forwarded requests.” This is one of the places I check several times per week since it’s rewarding to see that most of the requests I forwarded got accepted by the intended recipient and that I was able to make a difference in the life of my trusted contacts.

  5. resume writing

    It’s always important to analyze the requests in LinkedIn. There are so many fake accounts and just random people who want to get something from you that if you don’t choose what to do with each one carefully, your profile will be full of annoying and unnecessary contacts


Leave a Reply