Language services and technology provider SDL announces the launch of a new tool, SDL Content Assistant, taking another step away from traditional translation services towards comprehensive content delivery management. Nimdzi analysts obtained an advanced briefing on this new tool and have performed a preliminary analysis of the features, which we have summarized below.
In today’s day and age, content is key for growing brand awareness and reaching end-users across the globe. However, the sheer overload of content also means that large swathes of information are either overlooked or hard to find. For organizations with multilingual content, these challenges are further amplified.
Organizations generating digital content (and yes, this applies to us at Nimdzi, too) are faced with a challenge of how to leverage material to reach their ever-growing audiences.
SDL Content Assistant uses artificial intelligence to address this need. It helps marketers analyze their data and cuts down on the time needed to find meaningful and engaging content, which can then be used to fuel their digital content strategy.
The main asset of the tool is that it saves time. Instead of spending hours sifting through stacks of material and trying to identify relevant highlights and sharable content, users can focus on what matters most – growing their company’s outreach.
Let’s take a quick look into what you may find inside SDL Content Assistant:
There are two tabs, the workbench and the assets tabs. The user can load assets into the tool, which will be used to suggest content, summaries, and blurbs.
Summaries are automatically generated with the help of AI and are adjustable in length. They are based on keywords the user has preselected. Keywords are suggested based on products, people, locations, and organizations. While these keywords are automatically identified by SDL Content Assistant, the user also has the option of adding new keywords.
Similarly, there is a section for blurbs, which generates snippets of content suitable for Twitter and other social media posts. Blurbs bring the most value when they are concise and factual. For example, SDL’s AI pulls content which often begins with “Most,” “Despite,” and “All.”
Just below, in the workbench, is the stats section, which pulls (you guessed it) stats and figures. And below that is a Tweets section, generated based on the topics the user has selected.
The workbench’s Scrathpad is a space to whip up ideas based on assets, and the tool also auto-suggest content as the user begins writing.
The Assets tab has dashboards with insight on the assets. These include a domain tab with a radar chart which displays which domains are most dominant in the content. SDL Content Assistant currently includes the following domains:
- Adult, Arts & Entertainment, Autos & Vehicles, Beauty & Fitness, Books & Literature, Business & Industrial, Computers & Electronics, Finance, Food & Drink, Games, Health, Hobbies & Leisure, Home & Garden, Internet & Telecom, Jobs & Education, Law & Government, News, Online Communities, People & Society, Pets & Animals, Real Estate, Reference, Science, Sensitive Subjects, Shopping, Sports, Travel
Other dashboards include: Sentence Length Distribution, Uncommon Words, Key Phrases, Entities, Languages, Document Length Distribution, MT Sustainability, and the Complexity dashboard.
The Complexity dashboard analyzes various characteristics of the content including lexical richness, vocabulary choice, and syllables per word.
The Entities dashboard displays all the organizations, persons, locations and events recognized by the tool, which would especially useful to marketers when tagging content.
Of particular interest to global companies, the MT Sustainability Dashboard analyzes how suitable the content would be for machine translation into a number of languages.
The tool is feature-rich. That being said, it is not a magical solution to all content suggestion needs. Despite the use of AI (or, arguably, because of the AI), it does have its limitations, and content creation is still quite manual. Here are a few things we noticed during a first trial:
- The summary is not a connected, logical text but rather a random selection of sentences in the form of bullet points. The points do not build on one another, meaning context might be lost and information misunderstood.
- The spotlight section only lists sentences that contain the preselected keywords. So, if such a sentence is followed by one that elaborates on the point but either doesn’t contain the keyword at all or replaces it with a pronoun, this essential context will be missed.
- In the “stats” section, statistics relevant to the keywords are highlighted. However, these cannot be imported into the scratchpad and users also cannot read the full sentence they’re contained in (only a small snippet). This means that users have to refer back to the original document to find the full sentence with the stats there.
Overall, the SDL Content Assistant is a useful tool for marketers and contains a number of features that make life easier and have the potential to increase efficiency. However, it does not fully eliminate the need for users to read the original documents or refer back to them. The tool mainly provides snippets of information without context.
Once the original document has been read, the SDL Content Assistant will make creating bite-sized, social-media-friendly content creation easier and faster.
The product is currently available for free for a limited amount of content, but the company plans to expand its offering through a freemium model. Price levels have not been disclosed.
Our first-look review is that the tool looks promising. SDL has made a wise strategic choice to release as a freemium model, allowing users to try before they buy, and we encourage our readers to experience it for themselves.
Based upon our analysis, we find that SDL Content Assistant is just that – an assistant (albeit a potentially powerful one). As with many AI-powered technologies, it serves to enhance and optimize, rather than replace the efforts of an experienced professional.
We are excited to see how this tool will develop and look forward to covering it further in our ongoing industry analysis.
