When I have published in a journal managed by Elsevier, I have had to fill out ‘Highlights’. According to Elsevier, these highlights are ‘bullet points that help increase the discoverability of your article via search engines.’
My problem with this is not only that we already have keywords for journal articles, but that I do not believe scientific publishing should aim to ‘increase the discoverability’ of individual studies. It is great that scientists can find your work, but I do not like the idea of SEO in the domain of publishing scientific articles.
Here is another point from Elsevier on ‘Highlights’: ‘Highlights have been proven to widen the reach of your work and help to ensure that your article is brought to the attention of interested colleagues, both inside and outside your usual research community.’ Proven? Where? They do not provide a single link to a study demonstrating this. I am not saying that I would be interested in seeing such a study, but if you say something is proven, you need to prove it.
If you have something important to say, say it in the abstract. That is also why I love these highlights from a study published in the Journal of Development Economics:
The first of the three highlights outline what is being done in the article. The second and the third simply note that everything you need to know is in the abstract. Scientific publishing should be simplified.