Back to Blog

Exposure of the biggest Myths of Content Automation

Reading Time 4 mins | February 11, 2020 | Written by: Saim Alkan

– Reading Time: 4 minutes

More and more companies in e-commerce generate high-quality SEO landing pages and unique automated product descriptions with the AX Semantics software and profit from the opportunities of Natural Language Generation (NLG) . Nevertheless, many people have doubts about content automation. These concerns include, that automatically generated content contains search terms that make no sense or that content is automatically translated without human review before publication.

Google is taking a clear stand on the issue of automated text generation and is completely open about the handling and evaluation of automated content. We explain the seven things that are rated negatively by the search engine and clarify to what extent the use of text automated with AX Semantics is affected. SEO Content that contains search terms but does not make sense to the reader.

1. SEO Content that contains search terms but does not make sense to the reader

With the software by AX Semantics, content and style quality can be ensured by the user himself over a large amount of text. Structured data such as product information sheets, report data, etc. are used as the information base for each text. How these are interpreted and evaluated, i.e. the configuration of rules, evaluations, statements and tonality is completely in the hands of the editors. The software offers the possibility to design texts in such a way that they have an added value for the reader.

2. Content that has been translated without human review before publication

AX Semantics does not include an automatic translation function. It is possible to generate texts in up to 110 languages because the software supports the grammatical characteristics of these languages. However, human translators are responsible for content translation. They adapt and localize the texts from the source language into the target language as required. The content logic can be adapted from the source language so that the effort required is significantly less than for the source language.

3. Content that is generated by automated processes, e.g. Markov chains

In Natural Language Generation, Markov chains are a statistical method of generating text. Our software does not work with statistical models, but focuses on editor-controlled content creation. This means that the problems of Markov chains do not occur while using AX Semantics.

4. Content generated with automated synonymization techniques

Synonyms and variances are not automatically generated by our software. You create them yourself as a user in our software. The software uses a text from your created variations. There are no statements which you have not configured before. Inferior or wrong synonymies do not occur in this way.

5. Content created by scraping Atom/RSS feeds or search results

AX Semantics does not use scraping of Atom/RSS feeds or search results to create content. Structured data such as product properties and report data form the information base for content. You can select formats such as CSV, JSON or Excel to upload your data.

Another use case that you can implement with our software is the automated newsstream. News articles from RSS feeds are used as an input base. However, the newsstream articles are not just reproduced but put into a new context by our software. This results a higher level of creation You can find more information about the function of our newsstream in the following blog article.

6. Content created by combining content from different websites

The AX software does not combine content from different websites. Instead, the tool works as follows: You as the user upload structured data that forms the information base for the texts. You then configure a set of rules that defines how the data is treated and interpreted by the software and language building blocks according to your taste and in your desired tonality. The software uses these to assemble a text.

The functionality of AX Semantics software is absolutely in the sense of Google when used correctly and does not break any of the rules. Of course, the software is only a tool, just like Microsoft Office is – so if you want to manipulate it, you have the theoretical possibility to do so – but not more or less than you can do with Microsoft Word. In addition, we regularly come across another doubt that results from the automation of content – it concerns the volume of text: We will discuss this below:

7. Uploading thousands of texts at the same time

We can also eliminate these doubts: Google does not punish you if you put a lot of content online at the same time. The pages of your e-commerce shop, for example, are not all indexed by Google at once. It takes time for Google to work its way through all the pages. So don’t let yourself be confused. Google will not punish you for this!

More than 500 companies successfully use AX Semantics for automated content generation

The online price comparison portal billiger.de has already generated 10,000 unique and high-quality descriptive texts for their product portfolio. Furthermore, MYTHERESA, a luxury fashion e-commerce company has achieved an increase in rankings of up to 20 positions within half a year after the start of automated text creation. See further successful customer stories about automated content writing for yourself on our website.

Now it’s your turn: Give automated content creation a try & see if human-machine collaboration works here for you. It can help you to optimize your brand and increase your success in the future!

Saim Alkan

Saim Rolf Alkan is Chief Executive Officer at AX Semantics and a pioneer in the field of automated content generation. After successfully working in content for years, he decided that businesses needed a better tool: one that would allow man and machine to work together to produce the volume of content needed to thrive in the digital age. Saim developed a content solution that generates high-quality texts from data in 110 languages for use in industries including e-commerce, publishing and finance. He is also a lecturer and speaker in the fields of online communication and "robot journalism" and has written several books and numerous articles.