It all started with the fact that I read the article “Longrod or longed. The texts of some length like most of the readers of” the Spark. The described results were obtained on the basis of the analysis of the articles “Habrahabr”, Geektimes (GT), Spark and vc.ru. A brief conclusion was the following. The main content, not its length.
I originally listened to this opinion, it is logical and statistically correct, but after a while realized that this study is not representative in the context of content marketing in General. It should be considered purely within the studied resources (and then, only then I noticed that only some have been analyzed 400 or 400 articles). In the first place “Habrahabr”, the GT and Spark is a specialized community with its rating system. With extensive experience of authorship on resource TM (published by “Thematic media”, which includes “.” and GT — Prim.
Ed.), can say a few sentences about some cataclysm dispersion can collect a huge number of views through voting and ranking of articles in the main belt. Strategy event-driven content marketing the actual not for everyone, especially if we are talking about B2B segment. Please correct me if Im wrong, but so brilliant counterparts, as Aviasales, in this respect, I have not heard.
The interest to the correct answer led me in the Western blogosphere. One of the most popular and large-scale studies published on the platform Medium at the end of 2013. Data were collected on the basis of a million articles. It is cited still.
The result was a graph of total reading time of all visitors to the average reading time of a post (length of post) a visitor. Seven minutes of time to read — very variable variable. To my surprise, many began to translate the value to a constant. For example, Buffer, Marketing Land, Snap, and other mistakenly interpretiruya these seven minutes 1600-1700 words.
At that time, Medium was 5 million posts, with the exception of articles with more than 1 million views and articles with more than 20 minutes the predicted reading time. Do not forget also about the availability and variety of types of images, video and other media files in the posts. Identified seven minutes is an indicator of what time you prefer to interact with the publication. This is an important result, but there I do not fully agree.
At the beginning of the Medium research analysts have tried to determine the optimal length of the content based on the number of article views. After that was said. We care less about clicks and more about actual reading.
The time spent is a more representative indicator. The total amount of reading time is one of the most important indicators, only publications. It shows the potential of advertising sites. Equally important decisivamente materials and engagement in reading.
I also asked a well-known media business representatives about which metrics are the most rates for their publications. Coverage — the most important indicator for advertisers and stable business model of the media. Other important metrics include engagement/retention and loyalty. You need to be sure that the attachment of users to your resource is so great that allows you to secure the name of your brand in their minds.
The most valuable indicator is the number of users regularly needing in your content. Not those who come once a month, and those who read you every day. It is important, how many people and how you can increase their number. Number of returns (sessions per month, for example), time on site, pageviews (over time, less and less important) and conversion into something (subscriptions to various services). There is still recirculate — actions/views after entering the material, decisivamente materials (highly dependent on external factors, but is usually compared with the average).
Please note, to the left of the graph is written AvgSesonds Spent per Visitor, but in the context of all the research we are talking about reading time, which is determined by the technology Medium. Scrolling, pauses, activity tabs, and so on. Also, in contrast to the graph summarizing the conclusion about the uniqueness of the seven minutes, here along the Y-axis shows the predicted time reading posts On the chart above shows the proper intersection of the predicted time to read and actual.
From which it follows that two-minute article (~485 words) have the highest engagement rates. Does involvement with a measure of decisivamente, we can only speculate. The probability of decisivamente much more if fakticheski reading greater than or equal to the expected. But in reality, you can carefully and for a long time to read 30% of the material, and then to doubt in its favor and have a quick go through the headlines until the end. In the study were not focus on the previous graph.
In a Medium visualized the ratio of the average time the user read to the forecast, and the median reading time. The median component of the final graph is about seven minutes. Please note how much has changed reading massive publications. Why analysts Medium decided to focus on the median?.
We often noticed by visitors who are inappropriately spent a lot of time reading. We have excluded the time from inactive tabs, but still account for users spending a few hours. I think this decision is wrong. Again, there are different types of materials. The study on Medium to Medium I have read for more than an hour.
Analyzing along with them and rechecking all the facts many times — because it is research analysts Medium. As a result, I became a visitor reading inappropriately long time and do not fall under the sample. But already eliminated a huge number of people who carefully read the publication to Pocket, Instapaper, Evernote, about which I forgot. The task of content marketing in the majority — registration and sales. And in this paths play an important role even one-time visits.
I want to emphasize. I didnt say that the recognition of the reader does not matter. But I can read about the ideal length of a blog post Buffer and start to use their product. I hope you understand what I mean.
The double-edged sword. I can also read the blog digital Studio and never use their services. But here, they have either incorrectly built the strategy of content marketing or, for example, their content marketing focuses on recruiting. Go back to Analytics. The penultimate graph in the study a Medium was the average number of visitors to the post.
I have already said that the median component along the Y-axis did not feel right, but it is also data. The graph shows that the four-minute articles on average gaining ~600 visits, and a seven-minute ~750. In the aspect of marketing and the entire business, it is important not to forget about ROI (return on investment). It is obvious that two four-minute articles potentially have more value and lower cost. It would be interesting to see this same chart, but below the Y-axis display the number of words.
Second, if not first popular conclusion, dated 2012. This figure shows the dependence of the length of the content in relation to search results. About how to do analysis does not say anything. I was looking for an answer.
Perhaps this is due to the increased amount of reference weight for a more voluminous publication, which is written in detail in these articles of 2009. We live in 2016. Search results depend on hundreds of different factors. The volume of the article is humiliating low value in this list.
Three years ago my collection of plugins for Sublime in English after the boom on Reddit was the third in the Google results. It gave 200-300 clicks per day for several months. Volume — 1000 words. But Im not a SEO professional, correct me if Im wrong. The objective of the study was the most unpopular.
A detailed analysis was subjected to more than a million posts, but the following table shows the results only for 489 thousand articles that do not contain videos and polls. This work was carried out in 2015, companies Moz and BuzzSumo and is described on 30 pages. Mind seems to be nothing, but my research is not only of interest to mathematical laws. I was chasing the “Golden section” in the context of content marketing. Based on the data from the table, we can say that three articles of 1000 words will be more effective and more profitable one with 5000 words.
Now compare the difference of the average values and medians. Moreover, the study said that about 50% of all articles analyzed only receive a few publications in social networks. The final very large numbers in the majority obtained by the popular media. That is why research Moz and BuzzSumo so different with the Medium. Now remember seven minutes Medium.
The most valuable currency in the world — its your time. The Financial Times, The Guardian, Hufington Post, TechCruch, The Next Web and others are famous brands. You trust them. You “pay more” for their information, but you are not willing to give more than seven minutes to unknown author.