What are autogenerated texts?
Autogenerated texts (also called machine-generated texts) are texts that are fully or partially written by machines. This is not just a distant future scenario, but a practice that has been going on for years. There are several tools available for this purpose, and while these tools are no match for skilled copywriters, they have reached a level of quality where they can no longer be ignored.
It takes a lot of time and money to write good texts - and if you have a large webshop with tens of thousands of product pages and category pages that all need a unique text, it's an almost insurmountable task. Therefore, there are good reasons to use auto-generated texts - but unfortunately, the results rarely live up to expectations.
We collaborate with several webshops that have tens of thousands of item numbers, yet we have not used autogenerated texts ourselves. For the sole reason that we can't vouch for the quality. Instead, our approach has been manual copywriting for selected pages - typically prioritized by the number of Google searches and the customer's profit margin on the individual products.
Google's attitude towards autogenerated texts
Google has a set of guidelines on autogenerated contentin which, not surprisingly, they state that publishing content that is intended to manipulate Google's search results can have negative consequences. According to Google, such content includes, among other things:
- texts that make no sense to the reader
- texts translated via an automated tool without human editing
- Texts generated via automated processes
- texts generated by copying RSS feeds or search results
- texts that are composed of content from different websites.
It goes without saying that texts should make sense to the reader and that copying content from RSS feeds, search results and other websites is a bad idea. On the other hand, points 2 and 3 are a gray area, both because it is very difficult for Google to identify this type of text and because the texts are no longer necessarily so bad that they need to be sanctioned.
Text spinning
A simple method for autogenerating texts is so-called text-spinning (in English called spinning article). The method simply works by a machine automatically creating a number of variations of an original text - by replacing and swapping words and phrases, among other things - whereby the variations of the text each appear as their own unique text.
Unfortunately, the result is rarely useful. The language is often poor or incomprehensible and the texts are often not varied enough. If you're tempted to copy texts from your competitors, put them through a text spinner and use them on your own website, don't!
The only case where text spinning can be useful is for product variations. For example, let's say you sell a smartphone cover in 10 different colors and each color variation has its own product page. You could write one product text and then use a text spinner to generate 10 unique text versions in which you simply replace the name of the color.
Natural Language Generation
A more advanced method for autogenerating texts is Natural Language Generation. The method is a combination of text spinning and a set of conditional rules that define how specific product information should automatically be used in the texts and how the texts should automatically change according to the product information.
An example of a conditional rule could be that if a product has wireless WiFi, the phrase '[product name] has wireless WiFi with a range of up to [range] meters' is added to the product description. The product name and range are dynamically replaced based on product data. Another example is that if a product is over a certain size, the adjectives 'big', 'huge' or 'XL size' are added to specific places in the product description.
The more conditional rules you define based on different product data, the better the texts will be. In practice, you typically create a template (i.e. a collection of rules) for each product category. For example, one template can be used to generate product descriptions for televisions, another for nails and screws, and a third for razors.
Natural Language Generation works best for product descriptions because they are typically very consistent. Category texts and blog posts should always be written by hand. Therefore, Natural Language Generation does not make copywriting redundant. It also requires a skilled copywriter to set up the templates that are the foundation for autogenerating good texts.
How to get started with Natural Language Generation
There are several tools available for Natural Language Generation. Among the most widely used are:
The advantage of tools like the above is not only that new texts can be generated automatically, but also that the autogenerated texts are consistent and have a uniform wording. It's also easy to translate the texts into other languages because you don't have to translate each text individually, but simply translate the templates used to generate the texts.
Despite the many benefits of Natural Language Generation, you need to think carefully before taking the plunge. It takes a lot of work to set up the templates that can generate good texts, and even though the tools have improved, they are far from being able to compete with good copywriters. In many cases, the quality of the texts is low and the texts are not varied enough.
