Ask.School is an AI-powered parent communication platform for UK schools that embeds directly on school websites and answers parent queries from approved school documents. The school website chatbot is becoming a familiar feature across UK education, and the technology has advanced rapidly. Schools that started with a basic FAQ widget or scripted chat tool now have the opportunity to offer something far more useful — a chatbot that understands context, handles follow-up questions, and draws answers from the school’s own documents and calendar.

Ask.School examines what parents genuinely expect when they interact with a chatbot on a school website, what the best solutions now offer, and what features a school website chatbot needs in order to be truly useful. Whether a school is considering its first chatbot or evaluating whether an existing tool is fit for purpose, the questions covered here should help school leaders make an informed decision.

For a broader look at how AI chatbots can reduce the administrative burden on school offices, see our guide on how AI chatbots reduce school office phone calls.

What do parents actually expect from a school website?

Parents visit school websites with specific, time-sensitive questions. They are not browsing for interest. They want an answer, and they want it quickly.

The Department for Education’s guidance on generative AI in education highlights that AI is most effective when it reduces administrative burden and provides information efficiently. For school websites, this means meeting parents where they are — on a phone, at the school gate, at 9pm — with answers they can act on immediately.

Common parent queries include:

  • Term dates and inset days — “When does half term start?”
  • Uniform requirements — “Do Year 7 students need a blazer?”
  • Absence reporting — “How do I report my child sick?”
  • Event information — “What time is the Year 4 assembly?”
  • School meals — “What is on the lunch menu this week?”
  • Policies — “What is the school’s behaviour policy?”
  • Contact details — “Who is the SENCo?”
  • Admissions — “How do I apply for a Reception place?”

These are not complex questions. The answers exist somewhere in the school’s documents, policies, letters and website pages. The problem is that parents cannot find them quickly enough — or at all. As we explored in our post on why schools need a better parent portal, the root cause is often that information is spread across too many systems for parents to navigate easily.

For a detailed breakdown of the most common queries, see our post on what parents actually search for on school websites.

Why do traditional FAQ pages fail parents?

Most school websites include a Frequently Asked Questions page. Some include several, organised by topic. On paper, this seems like a sensible approach: anticipate the most common questions, write answers, and publish them in one place.

In practice, FAQ pages have several fundamental problems.

They require parents to find the right page first

An FAQ page only works if parents know it exists and can navigate to it. On many school websites, the FAQ page is buried two or three clicks deep, often under a generic heading like “Information” or “Parents”. A parent in a hurry is unlikely to find it.

They only answer questions the school has anticipated

FAQ pages are static. They contain only the questions that someone at the school thought to include. If a parent has a question that was not anticipated — even a perfectly reasonable one — the FAQ page has nothing to offer.

They become outdated quickly

Schools are dynamic organisations. Term dates change, staff move on, policies are updated, events are rescheduled. An FAQ page requires manual updating every time something changes. In practice, this rarely happens with the frequency required, and parents lose trust in the accuracy of the information.

They cannot handle follow-up questions

When a parent reads an FAQ answer and has a natural follow-up question, the FAQ page cannot help. For example, a parent might read that school uniform is available from a specific supplier, then want to know whether second-hand uniform is available. Unless that exact follow-up was anticipated, the parent has to search elsewhere or call the office.

They do not understand context

FAQ pages treat every question in isolation. They cannot consider the time of year, the year group of the enquiring parent’s child, or any other contextual factor. A question about “sports day” in January should probably return information about next summer’s event, but an FAQ page has no way to make that distinction.

They offer no conversation

Parents often need to refine their question. They might start by asking about “after-school clubs” and then narrow it down to “after-school clubs for Year 3 on Tuesdays”. An FAQ page cannot engage in this kind of iterative dialogue. It presents a fixed list and leaves the parent to do the work of matching their need to the available answers.

What is the difference between a static FAQ and an AI chatbot?

The distinction between a traditional FAQ page and an AI-powered school website chatbot is not simply about presentation. It is a fundamentally different approach to information retrieval.

A static FAQ page is a document. It contains a fixed set of question-and-answer pairs, written by a person, and published on a web page. The parent reads through the list and tries to find the question that matches their need.

An AI chatbot is a conversational interface. The parent types or speaks a question in their own words, and the chatbot interprets the question, searches the school’s approved knowledge base, and returns a direct answer. If the parent needs clarification or has a follow-up, the chatbot can continue the conversation.

Here is how the two approaches compare across the features that matter most to parents and schools:

Feature Traditional FAQ Page AI-Powered Chatbot
How parents ask Scroll through a fixed list Type a question in their own words
Language understanding Exact match only — parent must use the school’s wording Natural language — understands synonyms, variations, and colloquialisms
Follow-up questions Not possible Maintains context across a conversation
Availability Always available, but static Always available, and responsive
Content freshness Requires manual updates by staff Automatically reflects updated documents
Number of topics Limited to what staff have written Covers every topic in the school’s knowledge base
Personalisation None Can tailor answers based on context (e.g. year group)
Multilingual support Requires separate translated pages Can respond in the parent’s language in real time
Accessibility Depends on page design Conversational interface works for parents with varying literacy levels
Safeguarding detection None Can detect and appropriately handle safeguarding concerns
Analytics Basic page views Detailed insights into what parents are asking
Maintenance burden High — every change requires manual editing Low — update the source document and the chatbot reflects it
Search functionality Ctrl+F on the page Semantic search across the entire knowledge base
Out-of-hours support Same static content at all hours Same quality of response at midnight as at midday

The difference becomes most apparent in real-world scenarios. Consider a parent who wants to know whether their child can wear trainers on a PE day. On a traditional FAQ page, they would need to find the uniform policy, then the PE kit section, then look for guidance on footwear. With an AI chatbot, they ask “Can my child wear trainers on PE day?” and receive a direct answer drawn from the school’s uniform policy.

Or consider a parent whose first language is not English. They might struggle to navigate a traditional FAQ page written in formal English. An AI chatbot can understand their question regardless of how it is phrased and respond in clear, simple language — or even in their first language.

What features should a school website chatbot have?

Not all chatbots are created equal. Many of the chatbot tools available to schools are simple scripted systems that present a decision tree: the user selects from a set of options, and the system follows a predetermined path to an answer. These are better than nothing, but they share many of the limitations of a static FAQ page.

A genuinely useful school website chatbot needs a specific set of features. Here is what school leaders should look for.

Does it understand natural language?

The most important feature of a school website chatbot is natural language understanding. Parents should be able to ask questions in their own words, without needing to use specific terminology or navigate a menu.

A parent might ask “What time does school finish?” or “When do the kids get out?” or “Pick-up time?” — all three are the same question, and the chatbot should understand all three.

Natural language understanding also means handling typos, abbreviations and incomplete sentences. Parents are often typing on mobile phones while juggling other tasks. The chatbot should be forgiving of imprecise input.

Does it draw answers from the school’s own documents?

A school website chatbot must answer from the school’s own approved sources: policies, handbooks, letters, calendars and other official documents — Ask.School’s chatbot documents guide explains how schools control which sources a chatbot draws from. It should not generate answers from general knowledge or make assumptions.

This is a critical distinction. A general-purpose AI tool like ChatGPT or Google Gemini can answer questions about schools in general terms, but it does not know the specific details of an individual school. A school website chatbot must be grounded in that school’s own information.

When a parent asks “What is the school’s attendance target?”, the chatbot should quote the figure from the school’s attendance policy — not provide a generic answer about national expectations.

Does it handle follow-up questions?

Conversations rarely consist of a single question. Parents typically start with a broad question and then narrow it down based on the answer they receive.

For example:

  • Parent: “What after-school clubs are available?”
  • Chatbot: Lists available clubs with days and times
  • Parent: “Which ones are for Year 5?”
  • Chatbot: Filters the list for Year 5
  • Parent: “Is there a cost for the coding club?”
  • Chatbot: Provides the fee and payment details

This kind of multi-turn conversation requires the chatbot to maintain context across messages. Each follow-up question builds on what came before. A chatbot that treats every message as a standalone query will frustrate parents quickly.

Does it stay current without manual updates?

One of the biggest problems with traditional FAQ pages is the maintenance burden. Every time a date changes, a policy is updated or a new event is announced, someone has to manually update the FAQ page.

A well-designed school website chatbot should draw its answers from the school’s live knowledge base. When the school updates a policy document, the chatbot should automatically reflect the new information. When a new newsletter is uploaded, the chatbot should be able to answer questions about its contents.

This does not mean the chatbot should have unrestricted access to everything. Schools should retain control over which documents are included in the chatbot’s knowledge base. But within that approved set, updates should be automatic.

Does it detect and handle safeguarding concerns?

Any tool that interacts with parents and potentially with children must include safeguarding provisions. The Generative AI Product Safety Standards published by the UK government are clear that AI tools used in educational settings must include safeguarding guardrails.

A school website chatbot should:

  • Detect safeguarding language — If a parent or child types something that indicates a safeguarding concern, the chatbot should recognise this and respond appropriately, directing them to the school’s designated safeguarding lead or relevant external services.
  • Never attempt to counsel or advise on safeguarding matters — The chatbot should signpost, not intervene. Safeguarding is a human responsibility.
  • Maintain an audit trail — Schools should be able to review conversations for safeguarding purposes if required.
  • Not anthropomorphise the AI — The chatbot should be clearly identified as an AI tool, not presented as a person. The AI Product Safety Standards are explicit about this requirement.

Safeguarding is not optional. It is a core requirement for any tool that operates within a school context, as outlined in Keeping Children Safe in Education (KCSIE).

Does it work on any device?

Parents access school websites from a wide range of devices. School website analytics consistently show that mobile devices account for the majority of parent visits — often 70 per cent or more. A school website chatbot must work seamlessly on mobile phones, tablets and desktop computers.

This means responsive design, fast loading times and an interface that is easy to use on a small screen. Parents should not need to pinch, zoom or scroll horizontally to interact with the chatbot.

Does it respect data protection requirements?

UK GDPR and the Data Protection Act 2018 apply to any tool that processes personal data. A school website chatbot should be designed to minimise data collection.

The ideal approach is to allow parents to use the chatbot without creating an account, without providing personal information and without being tracked. If a parent asks “What time does school start?”, there is no need to know who they are. The chatbot should provide the answer and not store any identifying information about the person who asked.

The AI Product Safety Standards require that personal data is not collected for commercial purposes such as model training. Schools should verify that any chatbot vendor complies with this requirement.

For a detailed examination of data protection requirements for AI tools in schools, see our guide on data protection and AI in schools.

Does it provide analytics on what parents are asking?

One of the most valuable secondary benefits of a school website chatbot is the data it generates about parent queries. Understanding what parents are asking — and what they are struggling to find — allows schools to improve their overall communication.

If the chatbot data shows that 30 per cent of queries are about term dates, the school knows it needs to make term dates more prominent on the website. If parents repeatedly ask about a policy that does not exist, the school knows there is a gap in its documentation.

Good analytics should show:

  • The most common questions asked
  • Questions the chatbot could not answer (indicating gaps in the knowledge base)
  • Peak usage times (indicating when parents are most active)
  • Trends over time (e.g. admissions queries increasing ahead of application deadlines)

This data is invaluable for school improvement, not just for the chatbot but for the school’s entire approach to parent communication.

How does a school website chatbot compare to a scripted chat widget?

It is important to distinguish between an AI-powered chatbot and a scripted chat widget. Many schools have experimented with scripted chat tools — sometimes called “decision tree” or “rule-based” chatbots — and found them disappointing. The distinction matters because the two tools work in fundamentally different ways.

Scripted chat widgets

A scripted chat widget follows a pre-defined conversation flow. The parent is presented with a set of options (e.g. “Admissions”, “Uniform”, “Term Dates”), selects one, and is taken down a predetermined path of further options until they reach an answer.

The limitations are similar to those of a traditional FAQ page:

  • Rigid structure — The parent must follow the script. If their question does not fit neatly into one of the predefined categories, they reach a dead end.
  • Limited vocabulary — The widget only understands the exact options it has been programmed with. It cannot interpret free-text questions.
  • High maintenance — Every new topic requires a new branch in the decision tree, manually created and tested by staff.
  • No follow-ups — Once the parent reaches an answer, the conversation effectively ends. There is no way to ask a clarifying question without starting over.
  • No learning — A scripted widget does not improve over time. It answers exactly the same way on day one as it does on day one thousand.

AI-powered chatbots

An AI-powered chatbot uses natural language processing to understand the parent’s question and retrieves the answer from the school’s knowledge base. The key differences are:

  • Free-text input — The parent types a question in their own words. There is no menu to navigate.
  • Semantic understanding — The chatbot understands meaning, not just keywords. “When do the kids break up?” and “What is the last day of term?” are understood as the same question.
  • Contextual follow-ups — The chatbot remembers what was discussed earlier in the conversation and can handle follow-up questions naturally.
  • Automatic updates — When the school’s documents are updated, the chatbot’s answers are updated too.
  • Broader coverage — The chatbot can answer any question that is covered in the school’s knowledge base, not just those that someone thought to include in a script.

Here is a side-by-side comparison:

Aspect Scripted Chat Widget AI-Powered Chatbot
Input method Menu selection Free-text question
Understanding Keyword matching Semantic understanding
Conversation flow Fixed decision tree Dynamic, context-aware
Coverage Only pre-scripted topics Entire knowledge base
Maintenance Manual scripting for every topic Upload documents, chatbot handles the rest
Follow-ups Start over from the beginning Continues the conversation naturally
Languages One language per script Multiple languages in real time
Setup time Days to weeks of scripting Hours — upload documents and go live

For schools that have already tried a scripted widget and been disappointed, the difference when switching to an AI-powered chatbot is significant.

What happens when the chatbot does not know the answer?

No chatbot will have an answer to every question. Parents will occasionally ask something that falls outside the school’s knowledge base, or they will ask about something that has not yet been documented. How the chatbot handles these situations is a measure of its quality.

A well-designed school website chatbot should:

  • Acknowledge the limitation honestly — Rather than guessing or providing a vague response, the chatbot should tell the parent that it does not have the answer to that specific question.
  • Suggest alternative routes — The chatbot should direct the parent to the school office, provide contact details, or suggest a relevant page on the school website where the answer might be found.
  • Log the unanswered question — Every question the chatbot cannot answer is valuable feedback. It tells the school what information is missing from the knowledge base and what parents need that the school has not yet documented.
  • Never fabricate an answer — This is critical. A chatbot that makes up an answer — a phenomenon sometimes called “hallucination” — can cause real problems for a school. If a parent is told the wrong start time, the wrong uniform requirement, or incorrect admissions information, the school’s credibility is damaged. The chatbot must only answer from approved sources.

The AI Product Safety Standards address this directly, requiring that AI tools provide responses grounded in approved content and do not generate misleading information.

How should a school website chatbot handle multilingual queries?

UK schools serve increasingly diverse communities. In many schools, a significant proportion of parents speak English as an additional language. A school website chatbot that only works in English is excluding some of the families who would benefit most from easier access to information.

An AI-powered chatbot can detect the language a parent is typing in and respond in that language. This happens in real time, without the school needing to translate any documents. The chatbot translates its answers on the fly, drawing from the English-language knowledge base but presenting the response in the parent’s language.

This is a significant advantage over traditional FAQ pages, which would require the school to create and maintain separate translated versions of every page — an impractical proposition for most schools.

Multilingual support is particularly important for:

  • Admissions queries — Parents who are new to the UK school system may need to understand unfamiliar processes and terminology.
  • Safeguarding information — All parents need to understand how to report a concern, regardless of their English proficiency.
  • Medical and dietary information — Accurate communication about allergies, medications and dietary requirements is essential for child safety.
  • Event information — Parents who struggle with English should not miss out on school events because they could not understand the invitation.

Schools with high proportions of EAL (English as an Additional Language) families often report that a multilingual chatbot transforms parental engagement by removing the language barrier that has historically prevented some families from accessing information.

How does a chatbot embed on a school website?

One of the practical questions school leaders ask is how a chatbot actually appears on the school website. The answer depends on the chatbot provider, but a well-designed solution should integrate seamlessly with the school’s existing website without requiring a redesign or a change of hosting provider.

Embedding methods

Most school website chatbots are embedded using a small piece of code — typically a JavaScript snippet — that the school adds to its website. This is similar to how schools embed Google Analytics, cookie consent banners or social media feeds.

The snippet creates a chat widget that appears on the school’s website, usually as a small icon in the bottom corner of the page. When a parent clicks the icon, the chatbot opens and they can start asking questions. The chatbot sharing and access guide covers the different ways schools can make their chatbot available.

This approach works with virtually any school website platform, including:

  • School website providers — Greenhouse, SchoolJotter, RM Unify, Cleverbox, e4education, PrimarySite
  • Content management systems — WordPress, Wix, Squarespace
  • Custom-built websites — Any website that allows the addition of a JavaScript snippet

The chatbot runs on the provider’s servers, not on the school’s website server. This means it does not slow down the school’s website or affect its performance.

Branding and customisation

A school website chatbot should look like it belongs on the school’s website. This means customisable colours, the school’s logo, and language that matches the school’s tone.

Parents should feel that they are interacting with their school, not with a third-party service. The chatbot should use the school’s name, refer to the school’s specific terminology (house names, year group labels, site names) and present itself as a school tool.

Mobile experience

As noted earlier, the majority of parents access school websites on mobile devices. The chatbot should be optimised for mobile use, with a clean interface that does not obscure the rest of the website and is easy to use with one hand.

The chatbot should load quickly on mobile connections and should not consume excessive data. Parents accessing the school website on a bus or in a waiting room should be able to get an answer within seconds.

What about schools that already have a website search function?

Many school websites include a search bar. This is better than no search at all, but website search has significant limitations compared to an AI chatbot.

How does website search compare to an AI chatbot?

Traditional website search works by matching keywords. If a parent searches for “PE kit”, the search engine returns every page on the website that contains the words “PE” and “kit”. This might include the uniform policy, a newsletter from last term, a governor meeting minutes document and a news article about a sports event. The parent is left to sift through the results and find the relevant answer.

The problems are compounded by the way school websites are structured. Important information is often embedded in PDF documents (prospectuses, policy documents, handbooks) that many website search functions cannot index. A parent searching for “attendance policy” may not find the attendance policy if it is a PDF download rather than a web page.

An AI chatbot does not just search — it understands. When a parent asks “What is the attendance policy?”, the chatbot does not return a list of pages. It reads the attendance policy, extracts the relevant information, and presents a clear, direct answer.

If the parent then asks “What happens if my child’s attendance falls below 90 per cent?”, the chatbot can find the specific section of the policy that addresses this question and present it clearly, without the parent needing to read the entire document.

This is particularly valuable for long policy documents. A behaviour policy might be 20 pages long. A parent who wants to know the consequences for a specific type of behaviour should not need to read all 20 pages. The chatbot can find the relevant section and present it in a concise, accessible format.

How does a school website chatbot handle sensitive topics?

Schools deal with sensitive topics regularly — safeguarding, exclusions, special educational needs, complaints, bereavement. A school website chatbot needs clear boundaries around how it handles these areas.

Topics the chatbot should handle

  • General safeguarding information — “Who is the designated safeguarding lead?” or “How do I report a concern about a child?” The chatbot should provide this information clearly and directly, as it is publicly available on the school’s website and in KCSIE.
  • SEND provision — General information about the school’s SEND offer, its accessibility plan and how to contact the SENCo. This is public information that parents are entitled to access easily.
  • Complaints procedure — How to make a complaint, who to contact, and the stages of the process. Again, this is information that should be readily available.

Topics the chatbot should not handle

  • Individual child matters — The chatbot should never discuss or provide information about individual children. If a parent asks “What is my child’s attendance percentage?”, the chatbot should direct them to the school office or the relevant parent portal, not attempt to answer.
  • Active safeguarding cases — If a parent’s query relates to an active safeguarding concern, the chatbot should immediately direct them to the appropriate person or service, not engage in discussion.
  • Staff conduct — Questions about individual members of staff should be directed to the school’s leadership team.
  • Legal or disciplinary matters — Questions about exclusions, legal proceedings or formal complaints should be handled by people, not by a chatbot.

The key principle is that a chatbot should handle information that is already public and accessible, and should direct parents to the appropriate human contact for anything that requires individual attention, professional judgement or confidentiality.

What does implementation look like in practice?

For school leaders considering a website chatbot, understanding the implementation process is important. A tool that takes weeks to set up and requires extensive technical knowledge will not work for most schools.

Step 1: Upload school documents

The school uploads its key documents to the chatbot’s knowledge base — the creating chatbots guide walks through the full setup process. These typically include:

  • School policies (behaviour, attendance, uniform, safeguarding, SEND)
  • Parent handbooks
  • Term date sheets
  • Newsletters and letters home
  • Prospectus
  • Admissions information
  • Staff contact lists
  • Curriculum overviews
  • Extracurricular activity schedules

The chatbot processes these documents and builds its knowledge base from them. The school retains full control over which documents are included.

Step 2: Connect calendar and feeds

Many schools use calendar feeds (iCal, Google Calendar) to manage events. Connecting these to the chatbot means parents can ask about upcoming events and receive accurate, real-time information.

Similarly, if the school uses an online menu system for school meals, this can be connected so parents can ask “What is for lunch today?” and receive the current day’s menu.

Step 3: Customise branding

The chatbot should be configured with the school’s colours, logo and name. The welcome message should be written in the school’s own voice. Some schools choose to name their chatbot (though the AI Product Safety Standards require that it is clearly identified as AI, not presented as a human). Schools can also configure prompt suggestions to help parents get started with common questions.

Step 4: Embed on the website

The school adds the provided code snippet to its website. This is a straightforward process that most school website providers support. For schools using managed website services, the provider can usually add the code on the school’s behalf.

Step 5: Test and refine

Before going live, it is worth testing the chatbot with a range of questions that parents commonly ask. This identifies any gaps in the knowledge base that need to be addressed. It also gives staff confidence in the chatbot’s responses before parents start using it.

Step 6: Launch and communicate

Parents need to know the chatbot is available. A brief mention in the school newsletter, a social media post and a prominent position on the school’s homepage are usually sufficient to drive adoption.

Schools that launch with clear communication about what the chatbot can do — and what it is not designed to do — tend to see the best uptake and the highest satisfaction from parents.

How should schools measure the success of a website chatbot?

Deploying a chatbot is not the end of the process. Schools should monitor its performance and use the data to improve both the chatbot and their wider communication strategy.

Key metrics

  • Usage volume — How many conversations are parents having with the chatbot each week? A steady increase indicates growing trust and adoption.
  • Resolution rate — What proportion of queries does the chatbot answer successfully, without the parent needing to contact the school office?
  • Unanswered queries — What questions is the chatbot unable to answer? These represent gaps in the knowledge base or topics the school has not yet documented.
  • Peak usage times — When are parents using the chatbot? Schools often discover that a significant proportion of queries come outside office hours, validating the chatbot’s availability.
  • Office call volume — Are routine phone calls to the office decreasing? This is the clearest indicator that the chatbot is meeting its primary objective. For practical strategies on reducing call volume, see our guide on how to reduce school office phone calls.
  • Parent satisfaction — Many chatbot tools include a simple feedback mechanism (e.g. “Was this answer helpful?”) that provides direct insight into parent satisfaction.

Continuous improvement

The data from a school website chatbot should feed back into the school’s communication processes. If parents are repeatedly asking about something that is not well documented, the school should create better documentation. If a policy change generates a spike in queries, the school should consider whether it communicated the change effectively.

Over time, the chatbot becomes not just a tool for answering questions but a feedback loop that helps the school understand what parents need and how well those needs are being met.

What questions should school leaders ask chatbot vendors?

For schools evaluating chatbot options, here is a checklist of questions to put to vendors. The answers will quickly distinguish between a basic scripted tool and a genuinely capable AI-powered solution.

Functionality

  1. Does the chatbot understand free-text questions in natural language?
  2. Can it handle follow-up questions within the same conversation?
  3. Does it draw answers exclusively from the school’s own approved documents?
  4. How does it handle questions it cannot answer?
  5. Does it support multiple languages?
  6. Can it be updated automatically when school documents change?

Safeguarding and compliance

  1. Does the chatbot include safeguarding detection?
  2. Is the chatbot clearly identified as AI, not presented as a person?
  3. Does the vendor comply with the Generative AI Product Safety Standards?
  4. Has a Data Protection Impact Assessment been completed?
  5. Where is conversation data stored?
  6. Is any data used for model training?
  7. Does the tool maintain an audit trail for safeguarding reviews?

Technical

  1. Does the chatbot work on all devices (mobile, tablet, desktop)?
  2. How is the chatbot embedded on the school website?
  3. Is it compatible with the school’s website provider?
  4. What is the setup time?
  5. Does the school need any technical expertise to manage it?

Data and analytics

  1. What analytics does the chatbot provide?
  2. Can the school see which questions are asked most frequently?
  3. Can the school see which questions the chatbot could not answer?
  4. Is the data exportable for school improvement reporting?

Cost and support

  1. What is the pricing model? (Per student, per school, flat fee?)
  2. Is there a trial period?
  3. What support is available during setup and beyond?
  4. Are there any additional costs for features like multilingual support or analytics?

Schools that ask these questions will quickly be able to assess whether a chatbot will genuinely improve parent communication or whether it is simply a more expensive version of the FAQ page they already have.

How does Ask.School approach the school website chatbot?

Ask.School was built specifically for UK schools, with every feature designed around how parents actually use school websites and how schools need to manage information.

The platform embeds on any school website as a branded chatbot that answers parent queries from the school’s own approved documents. Parents ask questions in plain English — or in their own language — and receive instant, accurate answers drawn exclusively from the school’s knowledge base. There is no account creation, no login, and no personal data collection.

Key features include:

  • Natural language understanding — Parents ask in their own words, including colloquial phrases and abbreviations
  • Multi-turn conversations — Follow-up questions are handled naturally, with full conversational context
  • Automatic knowledge base updates — When the school updates a document, the chatbot reflects the change
  • Safeguarding guardrails — Built around KCSIE and the Generative AI Product Safety Standards, with safeguarding language detection and appropriate signposting
  • Multilingual support — Real-time response in the parent’s language
  • No personal data collection — Parents use the chatbot without accounts or identifiers
  • Full audit trail — Conversations are logged for safeguarding review
  • Analytics dashboard — Schools can see what parents are asking, identify knowledge gaps, and track usage trends
  • Universal website compatibility — Works with any school website provider via a simple embed code

For schools already thinking about building a branded chatbot, our guide on how schools can create a branded AI chatbot covers the process in detail.

All data is hosted in the UK on ISO 27001-certified infrastructure. Ask.School meets all 14 requirements of the Generative AI Product Safety Standards, and conversation data is never used for model training.

What does the future of school website communication look like?

The school website is evolving from a static information repository to an interactive communication channel. The shift from FAQ pages to AI-powered chatbots is part of a broader trend towards meeting parents’ expectations for instant, accessible information.

Schools that adopt AI chatbots now are not just solving today’s communication problems. They are building the infrastructure for a more responsive, data-informed approach to parent engagement. The insights generated by chatbot analytics — what parents ask, when they ask it, and what information is missing — are a resource that static websites could never provide.

As the Department for Education continues to develop its guidance on AI in education, and as the regulatory framework around AI safety standards matures, schools that have already implemented compliant, well-designed chatbot tools will be well positioned. They will have the data, the processes and the parent trust that comes from doing this thoughtfully.

The question for school leaders is no longer whether to add a chatbot to the school website. It is whether the chatbot they have — or the one they are considering — actually does what parents need it to do.

Explore Ask.School’s parent-facing chatbot at ask.school.