Digital Skills Audits: Finding Gaps Before Buying New Software

Before buying a new system, find out which digital tasks your team needs to perform and what currently gets in the way. A software demonstration can make a process look simple while hiding gaps in confidence, access, instructions or basic workflow. A useful skills audit separates those problems so that training and purchasing decisions address the right issue.

This guide sets out a practical audit for a small team. It is a planning method, not an accredited assessment or a reason to rank staff publicly. Use it alongside a wider training needs analysis, especially where the proposed software changes responsibilities as well as screen layouts.

Describe the work before listing the software

Start with tasks and outcomes: find the correct customer record, update a stock entry, share a document with an authorised colleague or produce a weekly report. State what a correct result looks like and who is allowed to perform the action. Knowing the name of a spreadsheet package is not evidence that someone can check a formula or interpret its result.

Choose a manageable sample of recurring tasks and a few important exceptions. Ask staff where they repeatedly need help, where information is re-entered and where errors are difficult to correct. Include people working different shifts or away from a desk. Their experience may reveal access problems that an office-only review misses.

The Department for Education’s essential digital skills framework provides broad categories covering communication, information handling, transactions, problem-solving and safe online behaviour. Its 2026 standards update applies to England and includes technological changes such as AI. These can inform your questions, but a workplace audit must still reflect your actual tasks and systems. (DfE: essential digital skills framework; DfE: essential digital skills standards, 2026)

Separate skill, access and process problems

When someone cannot complete a task, investigate why. They may understand the process but lack the necessary permission. A slow device, inaccessible interface, unclear ownership or contradictory instructions can resemble a training gap. Buying another course will not repair those conditions.

Ask the person to explain the steps using a safe example. Observe where progress stops, then test your interpretation with them. Avoid treating one interrupted demonstration as a final judgement of ability. If a task is rarely performed, a clear job aid may be more useful than expecting everyone to memorise it.

Collect evidence proportionately

Combine a short self-assessment with practical examples and discussion. Confidence can help identify where someone wants support, but it is not a complete measure of competence. A confident user may still share a file incorrectly; a cautious colleague may perform the task accurately but need reassurance about an unfamiliar feature.

Use fictional records or an approved training environment. Do not ask staff to expose customer details, passwords or confidential work simply to demonstrate a skill. Explain the purpose of the audit, who will see individual results and how those results will inform support. Keep the record limited to what you need for that purpose.

Offer suitable ways to participate. A timed screen exercise may measure typing speed or access barriers more than the intended skill. Agree appropriate adjustments and let people explain their reasoning where that helps you assess the task fairly. Check with the responsible manager before using an audit for any purpose beyond the stated development exercise.

Use a small, clear skills matrix

Record each task against an agreed description such as needs an introduction, completes with support, completes independently or can explain the approved method to others. These are suggested working labels, not official qualification levels. Add a short evidence note and the date, because familiarity with a system can change.

Do not average everything into one digital score. A person may confidently prepare reports while needing help with secure sharing. Keep important safety or authorisation requirements visible rather than allowing strengths elsewhere to conceal them. The useful output is a specific action: practise setting the correct sharing permission, with a review next week.

For example, a fictional six-person team may show that four people can enter records correctly but none knows how to recover an accidental change. That finding calls for a recovery procedure and supervised practice, not necessarily six introductory courses. Confirm the system’s actual recovery options with its owner before designing the exercise.

Prioritise what must change before rollout

Group findings by their consequences for the planned change. Some skills are essential before someone uses the system independently. Others can develop during supported use. Some findings belong to the supplier, system administrator or process owner rather than the learner. Give each action an owner and a realistic completion date.

Turn learning needs into a staff development plan with protected practice time. A supplier demonstration can introduce features, but staff also need to rehearse your workflow, ask questions and correct mistakes safely. Agree who will provide help during the first live tasks and how issues will be escalated.

Use the findings in purchasing and review

Ask suppliers to demonstrate the tasks your audit identified, including exceptions and accessibility needs. Build those requirements into the training brief. Check what support, practice access and documentation are included in the quote, rather than accepting a generic promise of onboarding.

After training and initial use, review the same tasks with appropriate examples. Compare accuracy, support needs and avoidable rework, while acknowledging other changes that affect performance. Our guide to measuring training results explains how to avoid claiming that one intervention caused every improvement. Update the matrix when roles or systems change.

Common questions

Should everyone complete the same digital test?

A shared introduction can reveal common needs, but the practical tasks should reflect each role. Someone who only views a report does not need the same evidence as the person who builds it. Keep essential organisation-wide behaviours visible, then add role-specific tasks and suitable ways to demonstrate them.

Can I use staff confidence scores as the audit?

Use them as one input, not the whole assessment. Confidence can highlight anxiety, interest or previous experience, but it does not reliably show whether a task is completed correctly. Pair a short self-assessment with a discussion and a safe practical example, then agree the interpretation with the person involved.

What if the problem is the software rather than the person?

Record the barrier separately and send it to the appropriate system or process owner. Give a clear example of the task and where it fails, without unnecessary sensitive information. Training should explain an approved workable method; it should not normalise insecure workarounds for a system defect or missing permission.

Does the government framework certify our staff?

No. Using a published framework to organise questions does not award a qualification or establish that someone has passed a formal assessment. The framework and standards can help describe areas of learning. If a recognised qualification is required, investigate its actual provider, assessment and eligibility arrangements separately.

How can I involve people who rarely use a computer?

Start with the tasks their role actually requires and the equipment they can access. Offer a supported introduction and a suitable setting for practice. Do not assume that experience with a personal phone transfers automatically to a workplace system, or that a lack of computer use reflects unwillingness to learn.

Should the best digital user become the trainer?

Strong task performance is useful, but teaching also requires clear explanation, patience and knowledge of approved procedures. Ask whether the person wants the role and has time for it. Provide preparation and support, and check that informal shortcuts are not being passed on as the organisation’s standard method.

How often should we repeat the audit?

Choose review points around meaningful changes, such as a new system, role or recurring error pattern. You do not need to repeat every exercise whenever one feature changes. Revisit the affected tasks, check whether previous support worked and keep the evidence current enough to inform the next decision.

What should the final audit report contain?

Summarise the tasks examined, the evidence used and the actions required. Separate learning needs from access, equipment and process issues. Include owners, dates and a way to check improvement. Share individual detail only with those who need it for the agreed purpose; a public ranking is rarely useful for planning support.

Sources and further reading

• DfE: essential digital skills framework — broad skill categories for life and work.
• DfE: essential digital skills standards, 2026 — updated England standards and qualification context.