We see a lot of different fonts in the annual reports that we work with throughout the year.
Around 80% of those fonts work perfectly when the PDF is converted to xHTML and then merged with the XBRL layer.
The last 20% are turned into so-called Unicode during the conversion process.
Figures in Unicode are stored as graphic elements, which means they are not suited for HTML use. They still work, and they still look good, but that’s because they are stored as graphics. The minute they are replaced with the technical amounts from the Excel template, which is needed to create the XBRL layer, the HTML will render them as real figures which will change the font.
In most cases, our converter will be able to match the fonts without issue. When a font is changed to Unicode, we typically see four different scenarios:
- Most often the font looks exactly like it did before the replacement.
- In a few rare cases the font changes slightly after conversion. This is however barely noticeable to the naked eye.
- In some even rarer cases, the font changes so much that it becomes noticeable. While they are legible, it does take away from the overall experience.
- In the rarest of cases (which covers approximately 1-2%) the font simply does not work. It looks bad and it is barely legible, and, in these cases, there is no other option than changing the font.
Because the XBRL layer needs actual figures and not a reference to a graphical element somewhere in the code, there is no alternative to changing the font.
So, if you experience issues with your font, the best solution is often for your marketing department or design agency to find another font which matches but does not convert into Unicode.