An Analysis of 100 Inbound.org Submissions

Inbound.org Incoming screenshot

When I interviewed Rand Fishkin about Inbound.org – the community-curated inbound marketing resource – earlier this year, I found his answers to be very open, interesting and eye-opening. In particular though, this answer stood out the most for me:

Steve: What’s your view on self-promotion? Should people be afraid to submit their own posts (even if they are really good, ideal for the audience), or would you much rather see people only sharing other people’s content instead?

Rand: If you’re submitting 1/100 things you produce, that’s fine. If it’s closer to 1/10, that’s probably crossing a line. We don’t currently ban/remove for self-promotion or self-submissions, but we will ban accounts that consistently submit low-quality stuff (from anywhere).

In particular, it’s that mention of the “1/100″ ratio. For everyone 100 posts someone submits, 99 should be someone else’s material – not their own.

Now we’re an industry all about marketing and promotion and regardless of the industry, I think if anyone’s proud of what they’ve written and genuinely thinks that it can help others, there shouldn’t be any shame in sharing it. I think I’ve submitted close to 1/20 of my own stuff, either from SEOno, my company’s blog or in the form guest blog posts, but a) not excessively and b) only when I think it’s genuinely useful to people. Here’s an example (which flopped anyway and made me feel a little like this)!

As a daily visitor of Inbound.org who also regularly tucks into the Incoming page (rather than just sticking to the main page), I’ve noticed a fair few folk who do not follow this rule even close to 1/10, let alone 1/100. So I got curious…

At around 9pm (BST) on Thursday (25th October), I scraped the 100 latest incoming posts to analyse them.

Disclaimer: I’m not doing this to show anyone up, to discredit the site (which I adore!) or to highlight how Rand’s utopian view of the site does not line up with the reality. It’s purely scientific – I’m taking some data and analysing it. Treat it as a social or even a psychological experiment: i.e. here’s a site and this is what people are doing with it!

How many people self-submit?

So the first thing to check was how many people self-submit/self-promote their content. The answer? 38%. More than 1 in 3 out of the sample. Wow. Sorry Rand…

What counts as a self-submission? Some are less obvious than others, so I considered all of the following:

  • When the submitter is the owner of the blog,
  • When they’re the owner of another blog indirectly (e.g. “in association with…”)
  • When they’re the author of the post, even if it’s not on their own site (e.g. a guest blog post),
  • When they’re submitting the post of someone they work for/with,
  • When they’re submitting something from another website that still promotes them in some way (e.g. if it’s a SlideShare presentation on a talk that they’ve done, etc.)

Of course, it’s not simply a case of saying that it should be around 1%, as not all submitters are created equal (so to speak) and some people contribute more than others. I suppose it’s a bit like Dr Pete’s recent post on “X% of Queries” –  there’s other factors at play and therefore it can be interpreted a number of ways. But regardless, even if we were to say that 5-10% was a fairer figure, 38% is still considerably higher than that.

Here’s a few other random-ish observations…

Were self-submitters a certain type?

I definitely noticed that self-submitters were usually company/agency profiles. Although a few individuals self-submit (and likewise a few companies/agencies don’t), it seems the case that individuals using the site will use it as intended – sharing content that’s written by others – while companies/agencies will share their own.

Who was the most selfless submitter?

Who submitted the most posts that weren’t their own? The wonderful Aleyda Solis, who submitted 7 posts that weren’t her own. I think she deserves a link for that. :-)

Ed Fry came second (with 4). Ed recently became the site’s general manager, so it’s good that one of Inbound.org’s leading figures is using the site exactly as intended.

Speaking of which, a fair few SEOmoz individuals were in the sample, including Jen and Cyrus, all of them submitting non-SEOmoz material. However you could argue that with a site like that, they don’t need to self-submit because inevitably someone else is going to do it anyway – there’s less pressure on them to consider doing so. (Note: I’m certainly not saying that they’d be that way inclined anyway, but there’s certainly a difference.)

Who was the most selfish submitter?

Who submitted the most posts that were all theirs? I’m not going to name-and-shame, and luckily for me, it was a tie between a few people, so it doesn’t seem feasible to link to them all anyway (hah)! But there were a few people who’d submitted 2 of their own posts and just that – no one else’s.

I won’t say who, but in particular, one self-submitter has submitted 20 posts in their Inbound.org history and all 20 have been self-promotional. Now that is a bit sucky.

Ok, so while I don’t plan to name-and-shame, in the spirit of TAGFEE, I’m still willing to pass on the data. Want to see the spreadsheet with the full sample? You can view it here.

General hat tips

I wanted to end on a few notes of thanks…

Cheers to @Andrew_Isidoro of SEOFoSho for recommending a scraper tool for me to use in order to obtain the data from the site. While I used a different one in the end, I still wanted to say thanks.

Cheers also to @ir_emery, @Nonentity and @paulgailey/paulgailey.com for recommending Google Drive as the best/easiest way to share a spreadsheet online.

Uncategorised ,

22 comments


  1. Pingback: Analysis of 100 Inbound.org Submissions (including 38% self-submit!) - Inbound.org

  2. Thanks for this analysis Steve :)

    We’re actually just about to roll out a new version of the site with some handy new anti-spam features to fight the self-promoters. I’ve had to tweet a few accounts asking them to share more than just there own stuff – often they’ll end up being some of the most valuable members of the community since they suddenly start sharing all they’re reading and liking rather than just thinking of Inbound.org as a self-promotional channel.

    • Steve

      No problem. Thanks for not being annoyed about it…! Like I said though, I didn’t do it to show the site up or anything – I just thought it’d make for an interesting analysis.

      That’s a bold and brave approach – contacting the ‘offenders’ directly via Twitter – but I’m glad to hear that it’s working effectively. I guess a few people will realise the error of their ways as soon as it’s pointed out to them, as it could be the case that they don’t realise that they’re doing anything wrong, which – to an extent – is fair enough and certainly forgivable.

      Looking forward to seeing the new site! :-)

  3. This is an interesting post, Steve, but I wonder if 100 submissions is enough? It seems that there are a number of people who submit a lot more articles than their own (Jason Acidre comes to mind) who are not listed here. I’d love for someone to get the database of submissions and do an in-depth analysis of this.

    Ed, you have my email :-)

    • Ben

      John – Coming right up!

    • Steve

      Hi John. I agree. 100 probably doesn’t do it justice but given I didn’t have access to the raw data (I didn’t even think to ask actually – rookie mistake on my part?) and the fact that the link scraping tool also grabbed all the miscellaneous links (e.g. “Discuss”, etc.), sorting it out was still quite the undertaking.

      If it’s any consolation, I refused to do less than 100, even though I originally considered grabbing a much smaller sample (25 or 50). But when it comes to using percentages, it seems daft to me saying “38%” when it’s actually 19 out of 50, if that makes sense. Had to be 100 (at least).

      But like you say… getting hold of the database of submissions and doing an in-depth analysis? I look forward to seeing it! :-)

  4. Interesting post. But who makes up inbound? Is it the big names or is it the hundreds/thousands of unknown professionals? Then how are we going to police if these unknowns self promote? Its basically impossible unless someone actually checks it out and flags it as self promotion?

    And from what I’ve seen there is no benefit to doing this good karma type activity.

    I gave up on flagging junk on inbound as it didn’t seem to go anywhere? If you look at the incoming stream, there is tons of low quality posts and blatant self promotion.

    @Ed that’s cool that you’re reaching out and engaging them rather than banning them. Can’t wait to see this new version :) Stuff I have posted has been ‘mysteriously’ deleted and I wasn’t even affiliated with it in anyway. I just thought it was cool and they were giving away actual free good products that we would be interested in.

    • Steve

      Hi Kris, thanks for your comment – you raise some interesting points.

      When it comes to completely irrelevant spam submissions, the Flag button cannot be faulted IMO. However, when it’s one person who’s always submitting themselves but it’s still on-topic, it’s much trickier. I notified Casey (one of the mods) a while back when one guy was upvoting his own posts so that they’d make the main page – you could tell because he’d have 2 or 3 votes within minutes of submitting and all of his articles had at least 3 or 4. As the Flag button really only becomes effective when multiple people use it (I’m sure Ed can verify this claim?) then Flagging this guy didn’t seem to work as I think I was the only person doing so. In the end I think I had to notify Casey via Twitter directly to bring it to his attention. The guys don’t mind this – on the contrary, they welcome it. ;-)

      As for the ‘mystery’ deletes… I can’t say. Perhaps it was mistaken that you were affiliated somehow? I’d best not say as I’m not officially affiliated with Inbound.org (just a big fan), so I’m sure Ed, Rand et al will be able to shed more light onto the situation than I could. :-)

      • Yeah, some of the self-submissions, while on-topic, are a complete waste of time. There’s probably already some rockstar filtering happening anyway.

        For example, Rand and I submitted the same post around the same time but it showed up twice b/c the site has a dynamic URL. His upvotes crushed mine but the same would have happened, I think, if Jason or Aleyda had submitted it instead of Rand.

  5. how big is the inbound community?

    • Steve

      Hey Larry. Until this morning, I would’ve said “I have no idea,” but given Inbound.org’s redesign, the new members section seems to allow us to look at all members, not just the top ones. As I type this, I’m seeing that it goes as far back as page 183, showing nearly 9,000 members.

      That said, how does one define “community”? I say this because I wouldn’t be surprised if a lot of people are/were one-off sign-ups/visitors, so I guess they don’t really count. I’m sure Ed, Rand et al will be able to shed some light on a more accurate figure, based on those who visit and contribute regularly. My guess is that it’s in the 100s or low 1,000s.

  6. Interesting analysis Steve! I wonder in which direction this will evolve over time especially given the new site. Like John, I’d love too to see deeper analysis over more data and over time :-) I suggest looking at self-submission rate given the user influence (maybe on twitter).

    I’m curious which scrape you used in the end? Something such as Chrome web Scraper?
    Cheers!

    • Steve

      Hi Samuel, thanks for the comment.

      That’s a good suggestion! It sounds like John’s on the case though… maybe fire him an @mention and suggest it to him (just in case he doesn’t see this)?

      As for the link scrape tool, I used this one in the end. My friend Andrew recommended this one, but if I remember correctly, the format that the data was presented in was difficult to work with. The (old) Inbound.org Incoming page also had other links such as “Discuss”, etc., so tidying up the data took a while, although it was worth it in the end because some of the conclusions that could be drawn were really interesting. :-)

  7. Hey Steve,

    Great Analysis Steve!

    I think the links of Inbound.org members are now changed because of new design. For example: Aleyda’s link is now http://inbound.org/users/view/aleyda not http://inbound.org/member/aleyda and the links are broken.

  8. Nice one Steve!

    Great idea for a post. Would be cool to see what data John can dig up in a follow up.

    p.s thanks for the shoutout. :)

    • Steve

      Thanks Andrew. Likewise – thanks for your assistance!

      Well according to John, he “may do it today” (source) so we might not even have to wait that long…! :-)

  9. Pingback: Communitybait – Taking Egobait One Step Further

  10. My captcha today was “Tiger Shark” which is indeed a type of shark.

    Further to you ranking really high for “inbound.org” – I can’t believe I’ve never seen this post! This is awesome stuff!

    Not surprised that self submitters (um, sounds wrong but never mind) are agency types. I can say that we’re very encourged to share stuff other people in our agency have done. And more often than not, if B3 writes a post, someone in B3 will Inbound it (even if it isn’t the person that wrote it).

    Really cool – love this kind of stuff!

    • Steve

      Haha, that made me laugh! :-)

      Yeah… Some agencies are good with it though. I’ve noticed it a bit with B3 and a few other agencies (who I shan’t name) but in those instances they usually share colleagues (which I guess is better than sharing yourself), plus they do tend to genuinely share and also be genuinely shared in return.

      It’s when people only submit their own posts, and who wouldn’t necessary be shared under any other pretence. That’s what bugs me.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>