I work in this industry, a few scattered thoughts/explanations of how things work for the uninitiated.
These figures are actually pretty decent compared to the World Bank, who did a similar exercise in 2014 and found just under a third of their reports were _never_ downloaded [1].
The article discusses _only_ UN Secretariat reports, which, to my understanding, excludes most agencies (e.g., UNICEF, UNHCR etc.). We're looking at a very small sliver of the UN system here, i.e., there are many, many, many reports produced by the UN that aren't discussed.
I'm struggling to find the source report for the 1,100 figure, but I think this is probably a massive under-estimate from _within_ the Secretariat, because AFAIU, the Secretariat includes the Office for Coordination of Humanitarian Affairs (OCHA), who did >300 situation reports (sitreps) in the first half of this year.[2] I think it very unlikely sitreps constitute over half of the reports produced by the Secretariat.
Looking at the role of reports in the sector generally, it's best to think about them as a sort of parallel academic system, indeed many are produced alongside universities. Without my cynic hat on, they've got three audiences/uses, all very similar to traditional academic publishing: 1) getting press coverage, 2) informing activities/policy, 3) informing other reports.
Like academia, you're looking at a very diverse body of work, in terms of quality, usefulness and (importantly for the discussion) breadth of relevance. You have reports with genuine, absolute humanitarian necessity, (e.g., sitreps) [2]; you have ongoing annual tracking on a range of issues [3], matters of record [4], so much navel-gazing crap on the Sustainable Development Goals (SDGs)[5] which are the global goals for the UN.
A lot of it is turgid, obtuse, expensively-produced crap; some of it is the canonical take on a particular issue (e.g., climate change, migration from Northern Africa to Europe) and will be used by governments when developing their policy response. Some of it will only get used by a few niche NGOs when writing their proposals for their next years' work, or as a footnote in another report which will be read by a very limited audience. Some of it will be used by humanitarian agencies to ensure they're not all delivering aid to the same region.
It's hard to defend the UN sometimes, it's certainly very easy to criticise. All in all, I agree with the comments in the thread that note that these reports aren't for a general audience. I also agree that reader count isn't necessarily a good metric: a good report with good policy recommendations that's read and implemented by a small number of policy-makers beats the shit out of the thousandth report on SDG implementation that's read by a lot of actors because it's broad enough to be relevant to more people.
Not all are a particularly good use of funds and human effort, but the same could be said of a huge chunk of academic publishing. It's also (mostly) targeted at improving the world. I'd encourage anyone getting too pissy in the thread to consider the amount of resources tech industry invests in getting people to click ads, to con people into subscriptions, to squeeze customers, undermine labour, encourage addictive behaviours and sell us stuff we don't need.
It's a very different world to the tech scene: as flawed, as diverse, as occasionally brilliant, arguably more necessary, arguably less impactful. Frustrating and fascinating in equal measure.
[^1]: https://openknowledge.worldbank.org/entities/publication/fa4...
[^2]: https://reliefweb.int/updates?view=reports&advanced-search=%...
[^3]: https://www.un-ilibrary.org/content/books/9789211065923
[^4]: https://www.un-ilibrary.org/content/books/9789213589960
[^5]: https://www.un-ilibrary.org/content/periodicals/26181061