askVERA: Systematic Automated Easy Read Production

Every organisation that commits to Easy Read hits the same wall, and it is rarely a wall of intent. It is a wall of throughput. Policies get updated quarterly, service letters change monthly, consultations arrive with fixed deadlines, and each one needs an accessible version. Automated easy read production is the answer to that throughput problem: a workflow where conversion, image selection and quality checking happen in minutes, and human effort is spent only where judgement is needed.

This post is for the person who owns the output: the EDI lead, communications manager or engagement officer whose to-do list contains not “make a document accessible” but “make forty documents accessible by the end of the quarter”.

The production bottleneck nobody budgets for

Traditional Easy Read production is craft work. A writer simplifies the text sentence by sentence. Someone sources picture symbols, checks licensing, and lays out pages so the images sit beside the right lines. Then the draft goes out for review, comes back with changes, and the layout work starts again. Done externally, the process is typically measured in weeks per document. Done internally, it competes with every other demand on the communications team.

That model works when an organisation needs a handful of Easy Read documents a year. It collapses when the demand becomes continuous. A local council serving its Public Sector Equality Duty, an NHS trust updating patient letters, or a charity refreshing its service guides does not have one conversion to do. It has a rolling stream of them, plus a backlog of existing documents that have never been converted at all.

The result is familiar: a small set of polished Easy Read documents that were prioritised, and a much larger set of everyday communications that never made the cut. The bottleneck is not commitment. It is production capacity.

What automated easy read production changes

Automated easy read production does not remove people from the process. It removes the mechanical work that consumed their time: the first-draft simplification, the image hunting, the layout, the formatting. askVERA was built around exactly this division of labour.

The workflow runs in five steps:

1. Upload the source. PDF, Word document, a URL or pasted text. Whatever format the original lives in, it goes in as it is, with no pre-formatting required.
2. AI processing. askVERA rewrites the content into Easy Read conventions: short sentences, plain words, one idea at a time, paired with picture symbols in a large, clear layout.
3. Review and edit. A human reviewer works through the draft, adjusting wording, swapping images and confirming meaning. The person stays in control of every published word.
4. Compliance scoring. Each document is scored against the EU Easy Read Guidelines, with 80% or above indicating a good standard. The score turns quality from an opinion into a checkpoint.
5. Download and share. The finished document is exported and distributed through whatever channels the organisation already uses.

The image work deserves a specific mention, because in manual production it is often the slowest part. askVERA includes a built-in Easy Read image library that is auto-tagged, so the right symbols are suggested against the right sentences instead of being searched for one at a time.

Quality control when volume goes up

The standard worry about scaling any content process is that quality slides as volume rises. In a manual workflow that worry is justified, because quality depends entirely on whoever did the work that week. In an automated pipeline, two controls hold the line.

The first is the compliance score. Because every document is measured against the EU Easy Read Guidelines before it ships, a team can set a hard rule: nothing is published below 80%. That rule is enforceable whether the team produces five documents a month or fifty, and it gives managers a single number to monitor across the whole output.

The second is the review step itself. Automation produces the draft; it does not approve it. Keeping a named reviewer on every document means the organisation can answer the question “who checked this?” for any published item, which matters to public bodies with obligations under the Equality Act 2010.

A worked example: clearing a sixty-document backlog

Picture a housing and care provider with sixty tenant-facing documents flagged for Easy Read conversion: tenancy guides, repair request instructions, complaint procedures, safeguarding information. An external route would mean commissioning each document, waiting on drafts, and coordinating dozens of review cycles across many months.

Run through askVERA, the same programme looks different. The coordinator batches the documents by type, uploading a group each week. Drafts come back in minutes rather than weeks, so the constraint becomes review capacity, which is plannable: if each reviewer can check and sign off a handful of documents per week, the coordinator can put a realistic end date on the whole backlog and track progress against it. The compliance score gives a consistent quality bar across all sixty documents, even though several different reviewers share the load.

The point of the example is not speed for its own sake. It is that automated production makes the work schedulable. A backlog stops being an aspiration and becomes a plan with dates.

Costing a production pipeline

Manual Easy Read production is hard to budget because effort varies wildly by document. Automated production gives you unit economics you can plan around.

For a team running continuous production, the annual plan converts a lumpy, unpredictable cost into a fixed line in the budget, which is usually what finance asks for before approving a rolling accessibility programme.

Building the pipeline into business as usual

Organisations that sustain Easy Read output at volume tend to do four things:

• Assign an owner. One person controls the intake list, batching and deadlines, even if many people review.
• Convert at the source. New policies and letters go through conversion as part of publication, not as an afterthought when someone requests an accessible version.
• Set the quality gate once. Agree the minimum compliance score and the review sign-off rule, then apply them to everything.
• Plan by reviewer capacity. Since drafting is no longer the constraint, schedule volume around how many documents your reviewers can approve per week.

Teams working in more than one language can also draw on askVERA’s multiple language support, currently in beta, to extend the same pipeline beyond English rather than standing up a separate process. And the pattern here, prove a contained quick win, then scale it into business as usual, is the same automation journey askelie recommends across every process, applied to accessible communications.

Easy Read demand is not going to shrink. The organisations that cope are not the ones with the biggest teams; they are the ones that turned conversion from a craft into a pipeline. Automated easy read production is how that turn happens.

Related reading

• AI for Policy Document Analysis and Simplification
• Simplifying Complexity: AI Easy Read Conversion