How does Article Forge 2.5 content compare to human-written content?
This test was last run on November 12th, 2020 with Article Forge 2.5.Skip to the results
This case study expands on past tests which show that readers cannot tell that Article Forge content was written by a machine (read that case study here) by directly comparing Article Forge against content purchased from freelance writers.
Given that there is a large range in the quality and price of human-written content, we created a 5-star scoring system. In this scoring system, 1-star is the lowest quality, cheapest, human-written content and 5-star content is equivalent to an article published in the New York Times, or similar.
Our goal is to progressively benchmark against each content level as Article Forge improves.
Our most recent study compared Article Forge 2.5 to what we consider 1-star human-written content.
We paid 300 people through Mechanical Turk to read one Article Forge article and one human-written article. The readers were asked to choose which article they thought was better based on: overall quality, readability, general flow, syntax, accuracy, usefulness, and grammar.
The article topics were selected by randomly sampling over 2 million user keywords to create a diverse dataset across many industries and niches. The topics range from “the benefits of apple juice” to “tips for fly fishing” and cover both short tail (broad) and long tail (specific) keywords.
We then generated an article for each topic with Article Forge 2.5 and purchased an article about each topic from freelance writers (we paid $2.20 for each 300 word article which works out to $0.73 per 100 words).
When comparing Article Forge 2.5 articles directly to 1-star quality human-written articles, readers thought Article Forge articles were better 32.65% more often than the human-written articles. The results were statistically significant with a p-value of 0.0034.
Overall, 67.3% of all readers indicated they thought Article Forge content was at least as good as 1-star freelancer content.
Because the results indicate that readers clearly prefer Article Forge content over 1-star content, we can conclude that Article Forge 2.5 creates higher quality content than 1-star level human-written content.
While the purpose of this case study is to benchmark Article Forge 2.5’s content quality against human writers, there are two more factors to consider: price and turnaround time.
- The 1-star articles we purchased were approximately $2.20 each ($0.73 per 100 words) while Article Forge is $57/month for unlimited articles (and unlimited words). For anyone who has a strong content demand, Article Forge is orders of magnitude cheaper while offering higher quality deliverables.
- In general, turnaround time for the articles written by freelancers ranged anywhere from a few hours to over a week, while Article Forge takes about 60 seconds to generate each full article.
Therefore, when taking into account quality, price, and turnaround time, Article Forge 2.5 clearly outperforms 1-star content.
As we continue to improve Article Forge, we will continue to compare new versions against both 1-star and higher tiers of freelancer content.