A Blog Commenting Experiment: Automation vs. Manual

Deviled Ham imageIf you’re pro-white-hat SEO then you’re going to love this post…

If you’re a WordPress blog owner like me then no doubt you’ll also be sick of the amount of blog comment spam your blog receives. This blog gets about 100-150 spam comments per day and because Akismet isn’t 100% accurate – and I’ve had quite a few false-positives (genuine comments marked as spam) – I usually have to glance through them all manually. Ack!

Why do people blog comment spam? It’s usually the case with lazy, ignorant or opportunist SEOs who want to get a lot of links easily. Using something like ScrapeBox or XRumer (note: link to Wikipedia, not to them directly), black-hats can fire off hundreds of comments on blogs. If even a fraction go live – on blogs that don’t allow for moderation or by blog owners who think they’re genuine – then that’s a fair few links for little effort.

…Or is it?

The Experiment

A while back, a friend of mine (a fellow SEO) and I decided to conduct an experiment. We would both try to find blog commenting opportunities and also leave comments on blogs linking to the same site, but take two different approaches:

  • His approach was slightly grey-hat: he used ScrapeBox to find the opportunities, although he would then submit to them manually with your typical generic “great post,” “this is amazeballs” and “you make me weak at the knees, good sir” type comments.
  • My approach was 100% white-hat: I would find blogs manually, actually read the content of the posts and reply with a meaningful comment that actually aligns with and responds to the post.

Given the fact that my friend was using a semi-black-hat/grey-hat approach, you’ll have to forgive me – and I’m sure you’ll understand – why I’m not revealing his identity. There you go, I said “his” – I’ve narrowed it down 50% for you. :-)

Oh and we only had one hour (each) to do it. Game on!

The Results – Submission

The first phase was the scouting and submitting part of the process.

My ScrapeBoxing friend searched for blogs by keywords relevant to the site we were linking to. He found dozens of opportunities. However, as I mentioned previously, he still decided to submit to them manually. But hey, at least the first part was taken care of automatically, and sometimes finding the relevant blogs can be harder and more time-consuming than submitting to them.

By the time he was done, he had submitted comments on thirty blogs. As I said earlier though, they were fairly standard, generic comments. But still… thirty comments. That’s a lot in one hour – about one every two minutes.

Then it was my turn… Using an advanced Google search (probably something like [inurl:blog keyword] and set only to show pages from the past month), finding relevant blogs via my approach took longer and was more time-consuming. I found quite a few, sure, but with some of them, either they didn’t allow comments or I didn’t know how best to reply to the post, so inevitably I ended up visiting and reading more blog posts than actually commenting on them. I then took the time to carefully craft a response, something that the blog owner would surely appreciate and therefore publish.

How many did I manage? A paltry four – about one every fifteen minutes.

So, thirty vs… four. Ha. Here’s a graph to visually depict the difference:

Graph 1: SubmissionYeah, it’s not looking very good, is it?

But wait, there’s more…!

The Results – Approval

Submission is one thing, but what’s the point if they don’t even go live?

We waited a couple of weeks before revisiting all the blogs that we’d commented on, in order to allow ample time for them to be approved.

How many of the ScrapeBoxer’s comments went live, bearing in mind he’d submitted thirty? You ready? Wait for it…

One.

Yep, just one. Out of thirty. That’s a 3% success rate. Bahahaha! Sorry mate.

How many of mine went live? All four. A 100% success rate. Not only that, but two of the comments gained responses from other readers and/or the blog owners themselves, and – in one instance – I was also approached by one of the bloggers about a guest blog post opportunity.

Four comments may not sound incredible, but a) they encouraged interactions and led to other potential opportunities, and b) it still kicked the arse of the black-hat approach anyway, so there…!

Using the same scale as the previous graph, here’s the difference when it comes to the actual number of links acquired:

Graph 2: ApprovalWho knew the quality approach would actually lead to more quantity, too?

The Conclusion

It's a Spam Dandy imageDon’t be disheartened, white-hats. I for one never would’ve suspected that a black-hat-led approach such as this would’ve failed so miserably. In fact, even if there had been more successful, with more than one comment going live, would it have led to the guest blogging opportunity or the replies from other readers? Probably not.

If anything, this just goes to show the importance of understanding the difference between links submitted and links that actually go live. After all, saying you’ve submitted links to thirty different sites may sound impressive, but if very few actually go live then what’s the point?

Anyway… Hooray for white-hat, long live white-hat, etc. Let’s just enjoy this small victory.

[Image credits: deviled ham by Casey Hussein Bisson; “SPAM-dandy” (whatever the hell that means!) by Elvis Kennedy]

SEO , ,

16 comments


  1. Steve, this is awesome, son. I knew people can tell spammers from a mile away! Yep, it’s more work but totally worth it!

  2. Chris

    great post, this is amazeballs and you make me weak at the knees, good sir

    • Steve

      HA! (And yet I’ve approved it – the irony!) ;-)

      • Chris

        I reckon you could up your own comment volumes by at least 100% by better refining your advanced search operators & therefore

        1. uncover more sites on a topic you can easily write insightful comments on

        2. uncover newer posts with fewer comments as I find the referral traffic is higher from the first few comments.

  3. Really great post, Steve! (by the way it’s genuine comment :) Your example shows how important research and relevance are for SEO. I feel really inspired, Thanks!

  4. I like it…

    Since low quality blogs are the only ones that accept low quality comments my bet is the four comment links you got are better than the auto-approve 10 fold… Good study!

    • Steve

      Thanks Andy!

      I think you’re right there. Not only that, but I made sure to target recent-ish blogs. I think the automated approach would’ve picked up blogs from any time (including those that are years old), so those pages may not get re-crawled – and therefore the links not crawled – anyway.

  5. Good study Steve – I guess the real benefit from commenting on blogs are the relationships you build and what they eventually lead to – long term ROI.

    • Steve

      Hi Matt, that’s absolutely it. There may be other SEO considerations (not counting the whole nofollow thing, of course) – e.g. domain diversity; nofollow/dofollow ratio balance – but it’s most certainly more of a branding/exposure and relationship-building tactic than a quantity-driving one.

  6. Hey Steve, I stumbled across this amazeballs post (and site)!

    I totally agree with everything you say, and ‘doing it right’ we can connect with others and create opportunities that a scrabebox blast could never achieve. Comment spamming is of course a total waste of time.

    I would though question the methodology of your test. Why did your friend use scrapebox to find the blogs, but then manually write the useless comments? If he used spun comments within scrapebox itself he could have spammed about 10,000 blogs in the hour rather than 30 and got a lot more than 1 approved!

    I’m pretty sure he would have got more than your 4, although of course there would have been little/no SEO benefit and definitely no chance of building a realtionship with a relevant blogger.

    • Steve

      Hi Harvey. That’s a good question and a fair point! I’m not entirely sure why he took that approach, other than the fact that maybe he wanted to write the comments himself – the spun approach leads to errors and poorly-written comments.

      Another thing you just made me think of: the velocity of links. Getting up to 10k links all in one go? Probably a sure-fire way to set off a few alarm bells at Google…!

  7. James

    Hi Steve,

    Download a free plugin called ‘Animal Captcha’ – it contains a type of captcha that captcha services cannot bypass. Your spam comments will drop from 100+ per day to zero overnight.

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>