Access triage

Online consultation tools: how to compare them

How to compare GP online consultation and triage tools on patient access, workflow, clinical safety, data protection and reporting, with a scoring sheet.

Ava research teamPublished 7 min read

On this page
  1. Why the choice matters more than it used to
  2. Six groups of criteria
  3. What patients need from the form
  4. What navigators and clinicians need
  5. Reference calls with other practices
  6. Criteria in detail
  7. AI features and regulation
  8. Data protection basics
  9. A weighted scoring sheet
  10. Running a trial
  11. Switching from an existing tool
  12. Where policy is heading
  13. Frequently asked questions
  14. References

Key takeaways

  • Compare tools on how they change your workflow, not on feature lists: the same tool can work well in one practice and badly in another.
  • Score patient access first. A form that patients abandon pushes demand back onto the phone.
  • Ask for clinical safety and data protection documents early; any feature that assesses symptoms may be a regulated medical device[1].
  • Run a structured trial with real staff and patients before signing, and agree how you would leave if it does not work.

Why the choice matters more than it used to

An online consultation tool is now the main front door for a large share of patients. By 2025/26, 100% of GP practices offered digital messaging and 81% provided online consultations, according to NHS England[2]. Since October 2025, practices must keep online consultation systems open throughout core hours[3], so the tool is in constant use rather than an optional extra.

The tool shapes the work that follows. It decides what information arrives with each request, how requests are sorted, who sees them first and how quickly patients hear back. A poor fit adds work for navigators and clinicians every day. A good fit can make a total triage or hybrid model sustainable.

GP practices offering digital tools, 2025/26NHS England annual report
GP practices offering digital tools, 2025/26Comparison. Digital messaging: 100%; Online consultations: 81%.100%Digital messagingDigital messaging: 100%81%Online consultationsOnline consultations: 81%GP practices offering digital tools, 2025/26Comparison. Digital messaging: 100%; Online consultations: 81%.100%Digital messagingDigital messaging: 100%81%Online consultationsOnline consultations: 81%

Source: NHS England (via GOV.UK), 2026[2]

Show the data
ItemValue
Digital messaging100%
Online consultations81%

Six groups of criteria

  • Patient access: can patients find the form, complete it on a phone and understand what happens next?
  • Workflow: does it fit how your practice sorts and responds to requests, or will you have to change to suit it?
  • Clinical safety: how does it capture red flags, and what assurance does the supplier provide?
  • Data protection and governance: where is data held, who can access it and what documents does the supplier provide?
  • Integration: does it work with your clinical system, telephony and the NHS App?
  • Reporting: can you see demand, response times and routing without manual counting?

Weight each group according to your priorities. A practice moving to total triage will weight workflow and reporting heavily; a practice with many older patients may weight access and phone integration.

What patients need from the form

Start the evaluation from the patient's side, because that is where most tools differ. A patient on a phone, perhaps unwell or anxious, should be able to submit a clear request in a few minutes. Watch for long identity checks, repeated questions, screens that do not fit a small display and medical terms the patient has to interpret.

The route into the form matters as much as the form. In 2025/26, about 2.4 million online requests a month started in the NHS App[4], and patients using the NHS App were more likely than users of other websites or apps to report a good contact experience[5]. A tool that works smoothly from the NHS App and from your own website, with the same wording in both, starts with an advantage.

Ask each supplier what happens after submission. Patients should get an immediate acknowledgement with a realistic response time, and replies should arrive in a way they will notice, such as a text with a link.

What navigators and clinicians need

The staff side decides whether the tool saves time. Navigators need one list of all requests, phone and online, with filters by type, age and urgency and a clear view of what is waiting longest. They need quick ways to reply, request more information, book an appointment or send a link without switching systems.

Clinicians need the request, the patient's recent history and any attachments on one screen, plus the ability to message, call or book from there. Every extra click repeated across hundreds of requests a week adds up to hours. During demos, time how long it takes to handle a typical request from opening to closing, and compare tools on that figure rather than on the number of features.

Reference calls with other practices

Ask each supplier for two practices of a similar size and model to yours, and call them without the supplier present.

  • How long did go-live take, and what went wrong?
  • How do navigators and clinicians feel about it after six months?
  • What do patients complain about?
  • How responsive is the supplier when something breaks?
  • What would you do differently if you were choosing again?
  • Have you changed your workflow to suit the tool, and was that a good change?

Criteria in detail

Questions to ask and evidence to request from each supplier
CriterionQuestions to askEvidence to request
Patient completionHow long does a typical request take on a phone? What share of started forms are submitted?Completion data from comparable practices; a demo on a phone
AccessibilityDoes it work with screen readers and in other languages? Can staff submit on a patient's behalf?Accessibility statement; staff-submission demo
NHS App accessCan patients reach the form from the NHS App with the right practice settings?Live example in the NHS App
Sorting and routingCan requests be routed automatically by type to admin, pharmacy, nurse or GP lists?Configuration demo using your request types
Red flagsHow does the form identify possible emergencies and what does the patient see?Clinical safety case and hazard log summary
Clinician viewCan clinicians see history, previous requests and attachments in one screen?Clinician demo with realistic data
Messaging backCan staff reply, request photos, send links and book from the same screen?Demo of the full response flow
Clinical system integrationDoes it write to the patient record, and what is coded?Integration documentation
Telephony linkCan phone requests be entered into the same list?Demo of navigator entering a phone request
ReportingRequests by hour, time to first review, routing outcomes and backlog in real time?Sample reports
Data protectionWhere is data stored, retention periods, sub-processors?Data processing agreement; DPIA template
AssuranceDoes the supplier meet NHS assessment standards such as DTAC and the clinical risk management standards?Assessment certificates and summaries
SupportTraining, go-live help and response times for faults?Service levels in the contract

Keep the same questions for every supplier so answers can be compared.

AI features and regulation

Many tools now offer automated features: suggested priority, summaries of free text, or routing based on what the patient wrote. These can save time, but they raise specific questions.

  • Regulatory status: the MHRA says many software and AI products used in health and social care are regulated in the UK as medical devices[1]. A feature that suggests urgency or clinical priority is more likely to fall into that category than one that sorts admin requests. Ask the supplier how each feature is classified.
  • Safety evidence: ask how the feature was tested, what errors it makes and how clinicians stay in control of decisions.
  • Data use: ask whether patient data is used to train or improve the system, and on what basis.

GPs are cautious for good reasons. Among GPs not using AI tools, 82% were worried about patient privacy and data security[6] and 89% about professional liability and medico-legal issues[7]. Bring those concerns into the evaluation rather than leaving them for after purchase.

Data protection basics

Online consultation requests contain health information, which is special category data under UK GDPR Article 9[8]. Before signing, make sure you have a data processing agreement, a completed data protection impact assessment, clear retention rules and a list of any sub-processors. Check how patients are identified, how attachments such as photos are stored and how requests from carers on a patient's behalf are handled.

A weighted scoring sheet

Example, fictional: a practice compared three tools using weights agreed by the partners.

Example weighted scores (1 to 5 per criterion group, fictional)
Criteria groupWeightTool ATool BTool C
Patient access25435
Workflow fit25543
Clinical safety20444
Data protection10454
Integration10353
Reporting10432
Weighted total100410385375

The scores came from a demo for each tool, reference calls to two practices using it and a two-week staff trial of the top two. Tool A scored highest because its routing matched the practice's navigation protocols, even though Tool C was easier for patients. The practice asked Tool A's supplier to simplify two form screens before signing.

Running a trial

  1. Define success

    Agree three or four measures before the trial, such as completion rate, time to first review and navigator time per request.

  2. Use real request types

    Configure the trial with your own request types and routing rules, not the supplier's defaults.

  3. Involve every role

    Navigators, admin staff, GPs, nurses and pharmacists should each use it for real work during the trial.

  4. Test with patients

    Ask patient participation group members to submit test requests on their own phones and report problems.

  5. Test failure

    Try a red flag request, a carer submission, a request with a photo and a request from a patient with limited English.

  6. Compare honestly

    Score the trial against the agreed measures and note anything that would need workarounds.

Switching from an existing tool

If you already have a tool, changing carries its own risks. Ask how historic requests will be kept and accessed, how long both systems will run side by side and how patients will be told. Check your current contract for notice periods. Plan the switch for a quieter time of year, avoiding winter and the weeks around bank holidays, and keep staff training close to the go-live date.

Look at what is driving the change. If the problem is a long review queue, a new tool may not fix it if the cause is review capacity. If the problem is that patients abandon the form, compare completion data carefully, because that is where tools differ most.

Where policy is heading

The government's 10 Year Health Plan commits to rolling out digital triage, ambient voice technology and the single patient record to cut clinical admin[9]. Expect national requirements and integration with the NHS App to keep developing. When comparing tools, ask suppliers how they plan to keep up with national changes and whether your contract covers those updates.

Frequently asked questions

What is the most important criterion when comparing online consultation tools?

Fit with your workflow and patient access usually matter most. A tool that patients find hard to complete, or that does not match how your practice routes requests, creates extra work every day.

Are AI triage features regulated?

Some may be. The MHRA says many health software and AI products are regulated as medical devices. Ask suppliers how each feature is classified and what safety evidence they hold.

How long should a trial run?

Long enough to cover a Monday and a full week of normal demand, often two to four weeks, with agreed measures set before it starts.

Can staff enter phone requests into the online tool?

Most tools allow it, and it is good practice, because it keeps phone and online requests in one list and treats them by the same rules.

References

  1. GOV.UK (Medicines and Healthcare products Regulatory Agency) (2025). Software and artificial intelligence (AI) as a medical device (opens in a new tab) ↑

  2. NHS England (via GOV.UK) (2026). NHS England annual report and accounts 2025 to 2026 (opens in a new tab) ↑

  3. British Medical Association (2025). Pressures in general practice data analysis (opens in a new tab) ↑

  4. NHS England (via GOV.UK) (2026). NHS England annual report and accounts 2025 to 2026 (opens in a new tab) ↑

  5. Ipsos (GP Patient Survey provider for NHS England) (2026). 2026 GP Patient Survey results released (opens in a new tab) ↑

  6. Nuffield Trust (2025). How are GPs using AI? Insights from the front line (opens in a new tab) ↑

  7. Nuffield Trust (2025). How are GPs using AI? Insights from the front line (opens in a new tab) ↑

  8. legislation.gov.uk (The National Archives) (2026). UK GDPR Article 9: Processing of special categories of personal data (opens in a new tab) ↑

  9. GOV.UK (Department of Health and Social Care) (2025). Fit for the future: 10 Year Health Plan for England (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo