Skip to content

FAQ

How accurate is the conversion? What if it makes mistakes?

Section titled “How accurate is the conversion? What if it makes mistakes?”

We automatically run an AI-driven visual comparison of the rendered HTML output against the PDF input. If the result doesn’t meet our standards for accuracy, we attempt to fix it. If the output can’t be fixed, we abort the conversion and notify you. It is possible to supply PDF input that can’t be reliably processed, but it’s rare.

Every conversion has an unconditional 30-day money-back guarantee. Click the feedback button at the bottom of any page and tell us what happened.

We list on the website and on every conversion exactly which WCAG requirements we test against. Every document has a 30-day, no-questions-asked money-back guarantee.

Every document is validated against WCAG 2.1 AA standards. We auto-fix common issues and flag anything requiring manual review. You can download a compliance report for each conversion.

What happens with complex layouts (multi-column, sidebars, footnotes)?

Section titled “What happens with complex layouts (multi-column, sidebars, footnotes)?”

We detect complex layouts visually and compare the output with the input, but our goal is accessibility. Where we have to sacrifice layout complexity for accessibility, we will.

Yes. Extracting and formatting equations from PDFs is a challenge. We automatically detect equations in the source PDF and use best-of-breed equation recognition to extract each one, format it properly, and insert it into the HTML.

How is this different from Adobe Acrobat’s auto-tag feature?

Section titled “How is this different from Adobe Acrobat’s auto-tag feature?”

Auto-tag creates tags inside a PDF. We create HTML.

HTML is the native language of the web and can be made more accessible than a PDF. If you convert a PDF to an accessible PDF, you still have a PDF. But if you convert it to accessible HTML, you have a live web page that can be viewed, updated, and exported to any format.

What types of files can I convert to accessible HTML?

Section titled “What types of files can I convert to accessible HTML?”

PDF is the most common input format, but we also support Word, PNG, JPEG, WebP, GIF, and TIFF.

Scanned PDFs are a special case. We detect a PDF created on a scanner that contains only pictures of documents rather than document content, and process it through a completely different workflow. Text on each page is extracted, images are analyzed and given appropriate alt text, and the images are recreated in the HTML with the alt text and any recognized text.

We can only process PDFs that are unlocked.

Yes. The process is independent of the language.

We produce WCAG-compliant HTML. Every conversion includes an analysis of what the document was tested against, along with any warnings or failures.

PDF uploads are currently limited to 10 MB.

It varies with the complexity of the page and the length of the document. A text-only document of 10 pages normally completes in a few seconds; a 100-page math document may take 10 minutes. You never have to wait for completion — we email you when it’s finished.

What do I do with the HTML output — how do I use it in my LMS?

Section titled “What do I do with the HTML output — how do I use it in my LMS?”

Either download the HTML and insert it into your LMS like any other file, or grab the share link and put it in the LMS.

Where are my documents stored? Are they deleted after conversion?

Section titled “Where are my documents stored? Are they deleted after conversion?”

Documents are stored on our servers, or our vendor’s servers, until you delete them using the dashboard.

The share link is designed to be public. Anyone you share the link with can access the documents.

No.

Yes. Just drop multiple files onto the Wizard or in the dashboard.

Of course. The API is documented on the website.

The Enterprise plan includes API access and webhooks.

The generated HTML can automatically be stored on any S3-compatible storage system. Self-hosted storage is available on Department and Enterprise plans only. See S3 bucket integrations for setup.

We use a combination of cloud-based and on-site servers with automatic failover to multiple providers. We distribute AI calls across at least three different providers to process tasks in parallel and isolate the system from failures in any one provider. It is very reliable.

Feel free to process ten pages at no cost.

Paper invoices are supported on the Enterprise plan.