Six years ago while I was working in a company outside academia, I had been involved in improving a process which was hot in academia. I have developed a new method which, among other notable advantages, improved all previous methods by deriving better results. Although some time passed, today it can still be considered as state of the art.

Despite I was not in academia, because of the quality of the results, I have managed to publish a paper about it in a top journal in that field. I hoped that the paper would get popular, and it would get cited many times, but today, after 4-5 years of being available, to my surprise, the work is still not cited. I see that new papers that come out and that are related, still refer to (lesser) methods, as though mine doesn't exist at all. As I am following the papers in that field, and despite a large number of related papers, I still don't see much improvement over my work, so it's pretty much current.

The paper uses proper keywords and wording (so it's easy to find), the journal is a good one (so people must have seen it), and it cites all other related papers (so I guess that relevant people in the field who have Google Scholar would be aware of it existence if they get notified of new citations). And still the paper didn't take off. The only reason I could think of is networking. Since at that time I was not in academia, I did not go to conferences to get the work (and my name) "advertised". But I refuse to believe that this was the critical factor.

The questions are: How to get cited and how to boost the impact of one's work? And did I do anything wrong?

I am now in my PhD, and I would like to learn how to boost the popularity of my upcoming papers.

  • 6
    "as mine doesn't exist at all." I think you meant to write "as though mine doesn't exist at all". – Faheem Mitha May 22 '14 at 12:35
  • 3
    Although I'm unsure of how prevalent this aspect is for what you are experiencing, some new ideas challenge the established framework in such a way that simply acknowledging them will not be in one's best interest for both current projects and future funding opportunities. – Mad Jack May 22 '14 at 13:53
  • 3
    My first thought on reading this is to cite yourself and those papers and demonstrate how your methods would have improved them and then try to get that paper published. If you're deemed correct by others in your field, you'll increase your number of cites by one, and perhaps more quickly earn the attention of the other authors. – Aaron Hall May 23 '14 at 19:55
  • 5
    I have managed to publish a paper about it in a top journal in that field — If this is a computer science paper, you made a tactical mistake by publishing in a journal directly instead of at a conference. To first order approximation, computer scientists don't read (and therefore don't cite) journals. – JeffE May 24 '14 at 16:54
  • 2
    this might be relevant (how to write a theory paper people will cite) eeb.cornell.edu/Ellner/CitedPapers.pdf – WetlabStudent May 26 '14 at 15:24

I can see a few reasons why your paper was not cited as much as you hoped it to be:

  • Networking indeed does go a long way towards being cited. In my experience this is especially true for areas where many competing approaches are being published (which, by the sound of it, is true in your case). Even if your paper is published in a good venue, this alone does not guarantee that you will find a critical mass of readers to kickstart the process. Citations are in my experience somewhat viral. Google Scholar, which for better or for worse, is used by many researchers for searching for literature, returns results ordered more or less by citation count. Hence, papers with a few citations on them are much easier to find on Google Scholar than those with 0 citations. Hence, you need to "bootstrap" a bit before citations start coming in more or less by themselves. The best (ethical) way to do this is to network - tell other researchers that might be interested in your paper about it
  • You should not per se exclude the possibility that you overestimate the value of your contribution. Maybe the things your approach was good at are not valued as highly by the other researchers. Maybe you underestimate the qualities of some papers that are cited over yours.
  • Even if your paper was really good, maybe this was to hard to see at first glance. In an ideal world, every researcher would strive to fully grok each related paper. However, in the real world (especially in busy fields), you have to assume that a reader will at first only casually browse over your paper (and only really read it if it seems very strong to him at first glance). If your paper does not seem strong at first glance, people will not cite it.
  • Maybe you are overestimating the value of the venue you published in. In many busy fields, papers published in a "good" venue (as opposed to "the best there is") already have a hard time getting cited. Further, check whether your paper is available with the standard subscriptions that most universities have (e.g., for CS that would be IEEE, Springer, ACM, ... depending on your field of course). There are some non-predatory publishers out there that are simply not available in the subscriptions of many libraries.
| improve this answer | |
  • 1
    Excellent answer. Many researchers tend to overlook the importance of advertising (call it "networking" if this use of the term offends you). There are tens of thousands of papers published each year, and it's easy for something important to be lost in the noise. Assuming your work is actually good, you'll have to sell your work loudly and convincingly for it to be adopted. – eykanal May 22 '14 at 13:55
  • 1
    To add to the last point: If the journal allows for it, publishing on a preprint (e.g., on Arχiv) may circumvent the problem of the paper not being available to some potential citers. – Wrzlprmft Aug 10 '14 at 11:12
  • 3
    @eykanal: "There are tens of thousands of papers published each year." Yes, about 250 tens of thousands in science alone, to be more precise: cdnsciencepub.com/blog/21st-century-science-overload.aspx. – Pete L. Clark Jul 17 '17 at 17:40

The first person who should cite your work is actually YOU. If you simply abandoned your work and you expected others to pick up on it, it mostly does not work this way, unless your work is really ground-breaking. When I search for something on a area I am interested in, it is easy to pick up papers with more citations (which you do not yet have) or authors who are more prolific in this area (which you are now not, since you only published this one paper). Also, people tend to be suspicious when they see "amazing" results, when they are not followed by further work on this area, since that might be some indication that those results were a bit "fishy" to begin with. In this sense, you are the one (at least initially) who should pick up your previous work, improve on it, compare with the newest methods and promote it.

If you asked some years ago, I would not believe in networking either. But the fact is that it works. If you go to some conferences and talk to some of your 'favorite' authors (who are also your main competitors) ask about their published work, talk about your published work (but DO NOT reveal unpublished work), perhaps ask to share methods, do a polite follow-up email soon, you will see that most people respond positively, not because you are sucking up to them but because they see that their work has actual impact. In most of the cases, they also want to improve their methods and see "the polite competition of today as the possible cooperator of tomorrow". So, understand that networking is not just PR but as appreciation to what others and you can offer to the scientific community.

| improve this answer | |
  • 5
    Excellent point about the fact that if you don't follow up on your own work, it might be considered by others "abandoned by the author", and thus not worthy of their consideration. – Floris May 23 '14 at 19:09

In many fields, the literature moves faster than most people can follow. A journal article can easily be missed. The chance of it being seen now that 5 years have passed is virtually none.

The best way to get your work known in many fields is to get out there and personally evangelize it. Give talks at conferences, talk it up to colleagues, etc. Getting someone else to start using your method is key.

| improve this answer | |

I'd just like to add, even if your methods and results were real game-changers, you don't necessarily have a good paper. Writing skills are critical for increasing your citation count. You might consider looking closely at highly cited papers, and see what the authors did differently. As others have noted, networking has a big role too..

Writing Science: How to Write Papers That Get Cited and Proposals That Get Funded is a good resource.

| improve this answer | |

If it is software related and there are no available source codes (or even executables at times!), or the code is very poor in ease of use, I will usually not cite it given an alternative.

Another way to put this in more general terms: to what extent have you made it easy to reproduce your research?

| improve this answer | |

Go to conferences, give talks on your work, highlight unexplored questions, extensions and applications. Work gets cited when you convince other people there's something interesting to do, and they then write follow up papers.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.