Document estates rarely become disorderly because one file is badly handled. They become disorderly because one file is copied, downloaded, shared, re-uploaded, renamed, and reused until nobody is fully certain which version still governs the public record. What looks like a minor publishing choice at the start often turns into a much larger management problem later.
That is the repeated-document problem. Once public content is treated primarily as a PDF file rather than as living web content, duplication becomes easier, stale copies become normal, and the original editable source can quietly disappear from day-to-day use. The result is not just inconvenience for readers. It is higher cost, weaker control, and greater uncertainty for the organisation itself.
Government guidance in both the UK and the United States points in the same direction for exactly this reason. GOV.UK says HTML is the most accessible format for publishing documents and should be the first choice whenever possible, while Section508.gov says PDFs are often not the most accessible or mobile-friendly option and that agencies should prioritise HTML and use PDFs only when necessary. That shared position matters because it reflects publishing reality, not fashion.
In the UK, GOV.UK guidance on publishing accessible documents says HTML should be the first choice whenever possible, and advises teams to turn PDFs into HTML if the content is intended to be viewed online. In the US, Section508.gov's PDF guidance makes the same practical point by telling agencies to choose formats that support accessibility, responsiveness, mobile-friendliness, and a digital-first user experience.
When one document becomes many
The cost of duplication is usually underestimated because it does not appear at once. A document is published on a website, saved in a shared folder, attached to an email, downloaded to a laptop, copied into a project workspace, and sometimes uploaded to another page by a different team. Each copy feels harmless on its own, but together they weaken the idea that there is one current version everyone can trust.
Once several copies are circulating, the website stops being the clear public source of truth. Instead, the organisation ends up managing a family of similar files that all look formal, all carry branding, and may no longer agree with one another. The burden then shifts from publishing content well to policing versions after the fact.
The UK evidence is direct on this point. The Government Digital Service has said that PDFs are less likely to be kept up to date, and that the problem becomes especially acute when the same document exists in multiple formats because every change has to be made everywhere. That is not a formatting inconvenience. It is a governance failure waiting to happen.
The US position supports the same broader conclusion, even where it is framed through accessibility rather than content operations. Section508.gov treats public electronic documents as governed content that must remain conformant, whether they are posted on a website, sent as an official communication, or included inside another digital service. Once copies spread, the difficulty of maintaining that standard increases quickly.
Government Digital Service guidance notes that users often download PDFs, keep them offline, and continue sharing them without returning to the website for the latest version. Section508.gov's overview of electronic documents makes clear that publicly available electronic documents remain part of the organisation's accessibility responsibilities, whether they live on a web page, appear as attachments, or circulate in other official channels.
Stale copies create operational risk
Stale copies are not merely untidy. They can leave readers with outdated instructions, old contact details, withdrawn policies, or superseded procedural information, all while the organisation believes the current version is already live somewhere else. The longer those copies remain in circulation, the harder it becomes to know where the practical risk now sits.
This is one reason PDF-first publishing can become expensive over time. A single update may require teams to identify every place where the file has been posted, stored, or sent, then decide which copies can be withdrawn and which cannot. That work is rarely budgeted properly because it arrives as repeated interruption rather than a single visible project.
The UK guidance explains the pattern clearly. GOV.UK says it is much easier to maintain one version of a publication in HTML than multiple versions in different formats, and GDS warns that PDFs are more likely to leave users with the wrong information because they are harder to update and more likely to remain in offline use. These are not theoretical risks, they are publishing behaviours observed at scale.
The US guidance strengthens the same case from a different angle. Section508.gov says agencies should use HTML as the default when creating and publishing content online, partly to avoid having to create and maintain separate versions, workarounds, or alternative solutions later. That is an accessibility statement, but it is also a sound document-management principle.
GOV.UK content design guidance states that HTML should be used wherever possible for read-only public content. Section508.gov's guidance on prototypes and pilots goes further by telling agencies to use HTML as the default when publishing content online, including forms, instead of relying on document formats such as PDF or DOCX.
Lost originals turn small changes into expensive work
The repeated-document problem becomes much worse when the editable original is no longer close at hand. In many organisations, the file on the website is not the true source, but only an exported output from a Word document, design file, or internal template sitting elsewhere.
Over time, the location of that source can be forgotten, access can be restricted, or the file can simply vanish into an old folder structure that nobody wants to touch.
Once the original is lost or detached from normal publishing practice, small revisions become disproportionately difficult. Teams are no longer maintaining living content. They are reconstructing a document chain from static files, then trying to decide which copy is authoritative enough to edit without introducing further inconsistency.
That is where cost escalates fastest. A small wording correction can become a mini-project involving retrieval, re-export, accessibility checking, approval, replacement, and re-publication, all before anyone addresses the possibility that older copies are still circulating somewhere else. HTML does not remove every management problem, but it keeps the organisation much closer to a single maintained source that can be updated directly.
The value of that model is reflected in government policy on both sides of the Atlantic. GOV.UK advises teams to turn PDFs into HTML where the content is meant to be viewed, and Section508.gov positions HTML as the default online format because it reduces the need to manage alternative versions later. In practice, that is also a hedge against losing source control over time.
Why HTML is the stronger public version
The strongest case for HTML in this context is practical rather than rhetorical. HTML keeps the public version closer to the organisation's maintained version, preserves website context, supports direct updating, and gives readers a clearer route to the latest information. It also makes it easier to avoid a sprawl of near-identical files that continue to travel long after they should have been retired.
That matters for accessibility, but it also matters for operational confidence. GOV.UK links HTML to easier maintenance, clearer updating, and better support for users' browser settings, while Section508.gov links HTML to accessibility, responsiveness, mobile-friendliness, and a digital-first user experience.
Those are different phrasings, but they describe the same underlying advantage: HTML behaves like managed content, whereas PDFs often behave like unmanaged files.
A PDF can still have a place where a fixed, downloadable copy is genuinely needed. What it should not usually be asked to do is carry the full burden of being the primary public version when content needs to remain current, usable, and under control. That is precisely the scenario where duplicates, stale copies, and lost originals do the most damage.
Practical conclusion
If an organisation wants fewer stale copies, fewer conflicting versions, and fewer document-control problems, the answer is not better file naming alone. The stronger answer is to reduce dependence on PDF as the primary publishing model for public information, and to treat structured HTML as the main version wherever possible. That keeps the source closer to the site, the update path shorter, and the risk of unmanaged duplication lower.
The repeated-document problem rarely looks dramatic at first, which is why it is often ignored for too long. By the time it becomes visible, the estate already contains duplicates, outdated files, broken ownership, and expensive uncertainty about what remains current. HTML-first publishing does not solve every content problem, but it gives organisations a much stronger foundation for preventing this one.
Sources
All sources are listed below with direct links.
- GOV.UK. Publishing accessible documents. https://www.gov.uk/guidance/publishing-accessible-documents
- GOV.UK. Content types. https://www.gov.uk/guidance/content-design/content-types
- Government Digital Service. Why GOV.UK content should be published in HTML and not PDF. https://gds.blog.gov.uk/2018/07/16/why-gov-uk-content-should-be-published-in-html-and-not-pdf/
- Section508.gov. Create Accessible PDFs. https://www.section508.gov/create/pdfs/
- Section508.gov. Conformance of ICT Prototypes and Pilots. https://www.section508.gov/test/conformance-of-prototypes-and-pilots/
- Section508.gov. Electronic Documents Overview. https://www.section508.gov/test/documents/