Your Impact Report is a PDF. Search Engines Barely See It.

I've downloaded and read through a handful of NGO impact reports recently, out of curiosity more than anything. Most of them followed the same pattern: a well-designed, 40 to 60-page PDF, genuinely strong work described inside it. But to a search engine or an AI system trying to find and cite that work, the document is close to invisible.

PDFs are technically indexable. In practice, they rarely compete

Google can index PDFs, that part isn't in dispute. But search engines tend to treat HTML pages as the priority format, and PDFs are structurally at a disadvantage: no crawlable heading hierarchy the way HTML has, no metadata search engines can read the way they read a title tag, and far weaker internal linking, since links inside a PDF get a fraction of the clicks a normal page link does.

The practical effect: an organisation's single best piece of evidence, the actual outcomes from actual programmes, ends up living in the format search engines are least equipped to parse and least likely to surface.

The real cost is friction, not just visibility

There's a second problem underneath the indexing one. Every time you make someone download a file before they can see what you accomplished, you're adding a step between curiosity and understanding. That gap has a name in UX research: interaction cost, the cumulative mental and physical effort someone spends before reaching value. A grantmaker skimming ten organisations in an afternoon is not going to open ten PDFs. They're going to read what's visible on the page in front of them, and move on.

What I'd actually try

Not abandoning the PDF entirely, some funders still expect one, and it has real value as an archival, printable document. The fix is narrower: pull your strongest 3 to 4 findings out of the PDF and publish them as their own short HTML pages or articles. Give each one a clear heading, a specific number, and a real sentence describing what changed. Keep the full PDF linked from there for anyone who wants the complete picture.

I don't know how much traffic this recovers for any specific organisation, that depends on too many variables to promise a number. What I do know is that right now, the best evidence most mission-driven orgs have is sitting in the one format that's hardest for anything automated to find.