Jump to content

I Created A Content Mill, Made 50 Shitty Articles, and Here's The Results


Neville

Let's put our "shady" hats on for a second.  What we're about to do isn't technically illegal, but it's slightly spammy:

We created a Content Mill to crank out tons of articles!

Content Mill Machine

 

Sometimes this year I decided I wanted to get 10,000 visits per day on KopywritingKourse whether or not I published an article or not.  This means the traffic would have to come through organic search (Google, Bing, Yahoo) and clicks from referring websites.

Why 10,000 visits per day?  Because it's a great stream of traffic that allows the business to grow without direct input.  Based on the current numbers, here's the projected email signups for KopywritingKourse if we hit 10,000 organic visits per day:

On low days 3% of organic visitors signup to the email list:

10,000 X .03 = 300 free email signups per day.

On high days 5% of organic visitors signup to the email list:

10,000 X .05 = 500 free email signups per day. 

Getting 300 - 500 people a day to subscribe to my email newsletter is great, especially if those customers come to me organically (which means FREE)!

If the pace of new articles and traffic continues, by the end of this year KopywritingKourse should theoretically hit 10,000 organic visitors per day with no changes.

HOWEVER, I wanted to figure out some ways to maybe "boost" the delta acceleration of the traffic. So I asked myself:

"How can I 10x this growth?"

Content mill growth acceleration

One of the conclusions I came to was building a Content Mill.

A "Content Mill" is a system designed to pump out a shit-ton of articles.  This could mean either a huge network of paid writers, or unpaid guest posters.

A content mill doesn't care if the articles are good, it just plays a number game to get articles ranked.

 

 

 

How A Content Mill Works:

Content Mill Article

Let's say a content mill can create 100 articles in a month, here's how it plays out:

  • 100 articles are published per month.
  • 10% of those articles "bubble to the top" and start ranking in the search engines or gain social traction.
  • 10 articles now rank well in the search engines or bring in organic clicks.

Now this may seem like a silly way to publish since 90% of the articles are failures, but if you think about what Forbes, Business Insider, and a bunch of other "large media brands" do....it's exactly this!!

Forbes recently put out a note essential tip-toeing around saying, "We need more eyeballs so we can sell more advertising!!"

"In the past few months there’s been a drastic move toward ad viewability — in other words, advertisers only paying for the ads we can prove that people see. In addition, advertisers are increasingly buying premium ads for new content, not old.

To keep pace with these changes we need a reset on the way we’ve paid our contributors. Starting April 1, 2016 we’ll pay the same rate we now pay per visitor to content that’s within 90 days of publication. We’ll pay 25% of that amount for visitors to content more than 90 days old."

Basically what they're saying is:

more articles content mill

Networks like this that have large "contributor networks" pump out 1,000's of articles PER DAY.

These brands will have something like 20,000 contributors required to publish 4 articles a week.  This equates to 80,000 articles a month, or over 2,500 articles per day.

Let's say only 10% of those articles break out of the mold and gain some traction, that's 250 articles per DAY that get traction!

Nevermind the fact that 90% of the articles suck ass, these large publishers still get 250+ articles a day from their contributor network alone that get a fair amount of publicity!!

 

This shows that creating a Content Mill isn't necessarily the most efficient way about creating hit articles, but it's definitely a model that has worked for SOME companies.

So.....let's try making our own Content Mill! (Kontent Mill)?

 

 

Let's make our Kontent Mill!

Content Mill Machine

As a side-experiment I decided to create a little Kontent Mill.

I knew for sure that 90% or more of the articles produced would suck at first.  But that's part of the game: Playing the numbers.

In 4 easy steps I had a content mill up and running that didn't require me to write anything at all:

 

 

STEP #1.) Target relevant traffic and keywords.

targeted articles

Getting 10,000 hits per day means NOTHING unless people are signing up for the email list or buying products from me. If it doesn't bring me money, then I don't care. This means I don't want to attract the WRONG CROWD.

If I write clickbait articles on recent celebrity divorces, that will surely get some clicks, but it won't bring the right audience.

So the traffic needs to be targeted to the fields of marketing, sales, and copywriting.

I paid someone on Fiverr.com to fish me up 50 articles related to writing and copywriting (kind of like I did here). I got back a big ole list of keywords:

Copywriting SEO Keywords

So I knew WHAT to write, but writing that many articles by myself?  YEESH!

 

 

 

STEP #2.) Get other people to write the 50 articles ($445.25).

Bossing Other Writers Arouns

I already got stuff to do bruh....so I can't be the one making all these articles! If I wanted the Kontent Milll to post 1,000 articles per year, there's no way I could do that myself. Therefore I cannot be the one initially making the articles.

Instead I hired someone to be the organizer of all the content, and gave them a blank check to go hire writers on Fiverr to do the writing.  I've used Fiverr before to do a myriad of different tasks include got a blog post written on Fiverr for only $5.50, and it worked quite well.

Checkout a portion of the orders we placed:

fiverr writing ordersWe tried to keep it cheap and get only $5 gigs. We used several freelancers multiple times, and sometimes ordered extra fast delivery. Total cost for articles was $445.25.

I ain't gonna lie, if I had to write all these boring-ass articles I would've blown my brains out.

Thankfully there's people out there who will trade a small amount of money to pump out an average to low quality post for $5.00, and they completed every single article for me!

HOORAY TO THROWING MONEY ($445.25) AT THE PROBLEM!

NOW......the quality of these posts kind of sucked balls.  And rightfully so!  We were paying people on Fiverr the lowest acceptable amount for each article, so you're not gonna get brilliant work done.

I was pleasantly surprised that not one article came back to us with plagiarism (we used this plagiarism checker for each article). I for sure thought 50% of these would come back positive for plagiarism and be unusable. They all came back legit-looking enough.

Onwards......

 

 

 

STEP #3.) Post the articles to a "Hidden" part of the site.

I didn't want the blog homepage to be cluttered up with enthralling posts such as "Prepositional Phrase Examples."

So we put them in a section of the KopywritingKourse site called "Writing Guides" and put it here:

https://copywritingcourse.com/guides/

Writing Guides

 

Here's all the articles we got made for the Writing Guides section:

We actually ran out of relevant articles to write, so we capped out around 38 articles in this section.

We then made another section called "The Glossary" where we defined different writing terms:

https://copywritingcourse.com/glossary/

We made 10 articles here before stopping.  A lot of the traffic from these articles would most likely come from high school students looking up a word.

These articles for the most part all suck.  BUT, that's the game we're playing here (and why I forewarned you this "sneaky experiment" to boost traffic is slightly spammy).

 

 

 

STEP #4.) Wait and see what "bubbles to the top."

content-bubbling-to-the-top

Any content mill expects that 90% or more of the articles will suck, but if 10% of the articles gain traction (either by social shares, website shares, or SEO rankings), then it was a success.

  • If you make 50 articles and 10% stick, that's 5 articles that worked.
  • If you make 100 articles and 10% stick, that's 10 articles that worked.
  • If you make 200 articles and 10% stick, that's 20 articles that worked.
  • If you make 300 articles and 10% stick, that's 30 articles that worked.
  • If you make 400 articles and 10% stick, that's 40 articles that worked.
  • If you make 500 articles and 10% stick, that's 50 articles that worked.
  • If you make 1,000 articles and 10% stick, that's 100 articles that worked.
  • If you make 2,000 articles and 10% stick, that's 200 articles that worked.

It often takes 6-12 months for an article to start "bubbling to the top."

For example, this humorous piece I wrote on how to scam people took about 3 months to start gaining some real traction in the organic rankings. Then after 6 months started to take off! It peaked around 9 or 12 months later.

This single post has brought in thousands of visitors, thousands of email signups, thousands of ranking search engine keywords.....all from one post that was free for me to write (although requires approximately 12 hours of work over several days):

Scam Post Lifetime Stats:

scam-post-stats

Everyday this one piece of content brings me several hundred new visitors.

Everyday this one piece of content brings me email signups.

Everyday this one piece of content keeps bringing in search engine traffic.

Everyday this one piece of content exposes my work to a new audience.

....all for free!

THAT is the power of a good piece of content.

However these Content Mill posts we cranked out were mostly huge pieces of crap.  Let's see how they did:

 

 

Results of the Content Mill Experiment:

 

Now let's take a look at the results of our shit-tastic articles we bought for $5.00 each:

Content Mill Guides Stats:

Guides Stats

These stats are horrifically bad!!!!

What's even worse is the bounce rates are SKY HIGH (often 100%) meaning every single person who visits the page from a search engine immediately leaves, and the email signup rates are pretty much 0%!!!  These stats suck suck suck!!!

Even posts I personally consider "not my best work" rack up far better numbers in a single day than all of these combined.

The "Glossary" we made has even more horrific stats.  They're so bad it's actually quite comical:

Glossary Stats:

glossary stats

Needless to say, so far this Content Mill idea has fallen flat, at least in the short term. I still think the idea can work, but it doesn't seem like something to invest more time in at this stage.

By publishing maybe 20 more good articles this year, we'll get to the 10,000/day mark just publishing at our normal pace.

Content Mills tend to work when a higher percentage of the articles gain traction.  In this case the traction rate was so low, and the quality low, it doesn't seem to be profitable to continue.

However for reputable brands like Forbes, their strong brand name allows them to get "guest contributors" that make some reasonably good articles.  Most of them suck, but at least 10% of the articles output are good.

As a much smaller company, it'll be harder for us to get the sheer scale Forbes can get.  Also, our business model isn't dependent on the sheer number of views the site gets.  This means a Content Mill might not be the best model to spend time and money on.

So it's been a VERY short amount of time, however I think there's a solid conclusion I've made for the content mill:

Stop spending time and money on it, and go back to creating few (yet higher quality) posts!

 

Mmmm. Did you like that?

If you liked this post describing how to start a Content Mill, consider downloading the whole post for your own files and joining the email list:

73HWFNWMb_Hjg1va60AlnCfeOaBOla224mvDYhc1

You'll get:

-- This whole post downloadable as a Google Doc, PDF, and Word Doc --

-- Weekly emails with awesome experiments like this --

-- Updates when a new post is live --

 

Hopefully you learned something and enjoyed this post!

Sincerely,

Neville Medhora - Copywriting Course Kurator

 

P.S. What do you think about big content mills?  Like em?  Hate em?  Worked for one? Got questions I can answer?


User Feedback

Recommended Comments



It's unreal how you got that much content (Kontent?) made for not that much money. Loved reading this and would be awesome if you post another update in a few months.

Thank you for all this info you put out.

Link to comment
Guest David

Posted

Dude milk your best work -> how to scam peeps part 2?
Link to comment
Guest Peter

Posted

I used to be a ghostwriter myself. And yes, I wrote for a content mill for a short while.

Maybe the problem lies with the what you asked them to write about. You were asking entry level writers to write content for a copywriting site...

If they knew enough to write about it, they wouldn't be entry level writers.

For other niches it may work a lot better - although it may still be a good idea to NOT use the content on your own site (hidden section or not). It can damage not only your site's rankings, but also your credibility. If you use it for satellite properties, you get the backlinks as well.

Just my 0.02c

Link to comment
Guest Hoo Kang

Posted

Nice write-up Nev.

That would be interesting to see what would happen if you paid $100 USD per article for higher quality long form content and tested that. So four of those compared to 38 not very good articles.

Link to comment
Next time a client INSISTS we churn out x articles a month I'm sending them here. Cheers Nev!
Link to comment
Guest Carlos Toro

Posted

It's great when you go into detail about how to create something (post/campaign/effort) that ends up being successful, as you have done in the past. In this case you pivoted and wrote about something that didn't work, which is even more valuable. It would be interesting to learn more about these "failure experiments". Great post!
Link to comment
Guest Adwet Harsora

Posted

"If you have the money to solve a problem, its not a problem." Its krazy that you were actually able to do this with just $445.25! I sincerely thought it would be a lot more than that when i started reading. Great Post Neville, Thank you for sharing.
Link to comment
Thanks bro, that was such a quality review of a failed test, and without opting in again (that's soooo fking irritating). Really appreciate you opening the kimono... again!
Link to comment
Guest Emils

Posted

Ohh man, $445.25 of solid research, hours of work, solid points backed by real-life experience.

This article itself is a perfect example of quality work and what it takes to get it done.

The very reason I believe why content mills exist - good work is not cheap, it takes time and people who can do this type of work are rare so they are playing the numbers game hoping to get lucky.

Do quality work and it will be x1000 times easier to stand out and get noticed.

Link to comment
Are you afraid that much shitty content is going to hurt your SEO on the rest of your site? Based on everything I've read, Google not only looks at the quality of individual pages but also of the entire domain.
Link to comment
Guest Krista @ MissCorporateDropout.com

Posted

First, I gotta say - your posts crack me up! I love how candid you are! I clicked on 2 of the articles in the writing guide, and they were of better quality than I expected. How long do you plan to keep the articles posted? I'd be interested to see if the results improve over the next 3 months or so.
Link to comment
Guest Robert

Posted

Great topic. Great article. With the availability of cheap work like Fiverr, it is really important to know what works and is worth it and, more importantly, what doesn't and is not.vvAgree that follow-up in a few months would be valuable.
Link to comment
Guest Allen

Posted

1. Lowering your domain's overall bounce rate probably is a bad idea. If they aren't punishing you for it today, they will some day.

2. Previously the value in a content mill was that every page on the internet had some small link value, and creating thousands and thousands of articles gave you a bunch of internal links that you could then point to better articles or your home page, boosting rankings and domain authority. The numbers are astronomical though. This takes enormous volume.

I believe this has been largely corrected and I've seen content mills delete tens of thousands of articles as a result. When google went after quality years ago they killed many of these mills.

3. Your articles won't rank because they don't have links supporting them and they're hidden on the site. Your site is telling google not to rank them as is the rest of the internet.

4. Putting this all together, to make this strategy work you'd want to put the articles on multiple external sites, so their poor bounce rates and generally terrible quality don't pull your own domain down. You then link to your site from these sites and use them not as a way to get traffic, but as a way to get "clean" links.

Usually these sites are propped up with not so clean links, acting as a filter between your "money site" and all kinds of sketchy stuff, but you could probably find ways to build links to these sites in a morally acceptable way. Even just sharing the posts of social media sites would help to get them moving.

At the end of the day though rankings are a function of links and shares and organic links and shares don't happen to low quality articles. It's not an easy game to win.

Link to comment

Pay for content? You are sitting on a gold mine of content that is paying you!

You should make part of the assignments of the kopywriting kourse that students have to submit articles for your site.

Paddle downstream.

Link to comment
Guest Jon Symons

Posted

The results are not surprising for such crappy content. However, I do think I see a flaw in the experiment. All of the articles are on the topic of "writing." While writing is related to Kopywriting ;) it isn't really. As you said the audience for these articles will be high-school students doing homework. These kiddles will ALWAYS have 97.3% bounce rates. Curious why you didn't get articles that would target your audience of people who want to improve their copywriting. I'd redo the experiment and get proper article topics written based on long-tail keywords from your own Analytics. Content mills need to be either highly targetted or click-bait based catering to the hoi polloi. In the middle, like yours will be doomed.
Link to comment
Guest Neville

Posted

Thanks Yuri! Unfortunately it was relatively low quality, so I'll have to go back and modify the one's that start bubbling to the top.

I think in 6 or 9 months I'll revisit this post with updated stats to see how it does.

Link to comment
Guest Neville

Posted

Thanks Peter!

I knew most of the articles would be crap, and didn't mind for this experiment. Content Mills try to see what bubbles to the top despite the low quality.

Then the second step is to re-vamp posts that are starting to gain traction and actually make them good. This way you already know you've got a winner, and by putting a little effort into it can get Top 3 rankings.

Link to comment
Guest Neville

Posted

I think that would actually work far better in terms of quality, HOWEVER the argument against it is that you make only a very narrow band of articles.

With cheap and shitty articles, I can cover 50+ keywords for cheap, see which ones gain traction DESPITE being shitty, and then update the articles that bubble to the top.

But yes, for $100/each you can get moderately ok articles written. They would be good filler articles, but generally not world-class articles.

Link to comment
Guest Neville

Posted

That's a perfect way to explain it to them Liam! Show them the stats of the entire content mill vs. ONE damn good piece of content (the scam post) stats!
Link to comment
Guest Neville

Posted

I was surprised also Adwet. I probably could've got better quality with clearer instructions, more communications, and more patients (we express-ordered some which added to the cost).

However even as is, it wasn't a TON of money for fully finished articles. Glad you liked!

Link to comment
Guest Neville

Posted

Glad you enjoyed and learned something Cris, appreciate it :)
Link to comment
Guest Neville

Posted

Thanks for the kind words Emils! I originally was just gonna write a post ABOUT content mills, but it sucked, so instead I just created one, and therefore this article was 100x more interesting from the concrete examples :)
Link to comment
Guest Neville

Posted

Not really. It's a debated subject, but while the articles are crappy, they are NOT spammy or plagiarized, so it's unlikley the overall domain will get penalized for it.

Keep in mind my personal blog Nevblog.com has thousands of posts that go nowhere, but it doesn't downgrade the quality of the posts that have traction.

Link to comment


×
×
  • Create New...
Guest