A chatbot that answers
from your own documentation.

AI chatbot development services for support and sales: assistants grounded in the docs, policies and past tickets you already have. Every answer shows its source, and anything uncertain goes to a human instead of a guess. Built on your helpdesk.

Get a Proposal See Pricing

Reply within four business hours (typical) · Human handover built in

Brands our teams have delivered for

What we do

PixelCrayons builds support and sales chatbots grounded in your own documentation, on the stack you already run. Answers cite their sources, and low-confidence questions escalate to a human by design. Guardrails and an evaluation suite ship before launch, not after. The first working pilot runs on your real docs. We work direct for brands and white-label for agencies.

Grounded in your documentationAnswers written only from your help centre, policies, manuals and past tickets, with the source linked.
Escalation to a human, by designRefunds outside policy, legal questions, angry customers and anything ambiguous go to your team.
Guardrails & evaluation before launchAn evaluation suite built from your real tickets runs before launch, and you sign off on the scores.

Related work · SaaS · UK

B2B SaaS: from invisible to answer-engine cited.

Read the case study

Selected work

Judge the work,
not the adjectives.

Outcomes tied to real engagements, not averages.

All case studies →

2004

Established

500+

Agencies served

4,500+

Projects delivered

340%

Revenue growth · 7 months

eCommerce

+127%

Organic traffic · 5 months

SaaS

85%

Faster delivery · backlog cleared

via agency partner

Checked 15 September 2026

Where the work happens

Answers grounded in your documentation
Sources shown on every answer
Human handover built into every assistant
Reply within four business hours (typical)
Clear IP & Ownership Terms
Evaluation before launch, logs after
Answers grounded in your documentation
Sources shown on every answer
Human handover built into every assistant
Reply within four business hours (typical)
Clear IP & Ownership Terms
Evaluation before launch, logs after

What it can actually answer

Every demo answers tier one.
You are buying tier three.

Chatbot questions come in four tiers, and each one costs something different to serve. The tier a vendor demos is almost never the tier that decides whether the project was worth doing. This is the ladder we scope against before anyone writes a prompt.

01
Answerable from your documents
Assistant
“What is your refund window?”
Requires
An index of the docs you already have
No systems touched, nothing to authenticate. This is the tier in every demo, and it is useful. It is just not the hard part.
Step change: everything below needs your systems
02
Needs this customer’s own record
Assistant
“When does my plan renew?”
Requires
Identity verification, then read access to the system of record
The real step change. You now need a login boundary the assistant respects, and a rule for what it does when identity is uncertain.
03
Changes something
Assistant, on confirmation
“Cancel my renewal before it bills.”
Requires
Write access, an audit trail, and a reversal path
Write actions get an explicit confirmation step and a logged record: what was changed, by whom and on whose instruction. A wrong write is not a bad answer. It is a support ticket with money attached.
04
Judgement, exception or an upset customer
Human
“This is the third time this has happened.”
Requires
A person, with the conversation already summarised
Handed over deliberately, not as a failure. The measure of this tier is how much context arrives with the customer: nobody should have to repeat themselves to a human after explaining it to a bot.

The jump in difficulty is not between tiers three and four. It is between one and two: the moment the assistant needs to know who it is talking to.

Illustrative tiers: question examples are generic, not client data

In every build

What a chatbot build
actually includes.

Grounded in your documentation

The chatbot answers from your own help centre, policies, manuals and past tickets. It finds the relevant passages first and writes the answer only from them; the industry calls this retrieval-augmented generation, or RAG, and it is the same layer under our knowledge systems. Every answer links its source. No source, no answer.

The foundation

Escalation to a human, by design

Confidence thresholds decide when the assistant stops. Refunds outside policy, legal questions, angry customers and anything ambiguous route to your team with the full conversation and the relevant document attached. A chatbot that never hands over isn't confident. It's reckless.

Every build

Guardrails & evaluation before launch

Before a single customer sees it, the assistant runs against an evaluation suite built from your real tickets: grounding checks, tone checks, resistance to users talking it out of its rules (prompt injection) and the topics it must refuse. You see the scores and sign off on launch. The suite re-runs whenever your docs change.

Before launch

On your stack, in your channels

The assistant plugs into the helpdesk, CRM and website you already run, and the channels your customers actually use, rather than forcing a platform migration. We stay model-agnostic too. The model underneath is chosen on fit, cost and data residency, and structured to be swappable without rework.

Included

Handover, logs and ownership

Every conversation is logged and auditable, so you can see exactly what was said, what source it used and why it escalated. At handover you get the runbook and admin training. Your materials remain yours. Bespoke deliverable rights, licences and handover are agreed before work starts. Updating the knowledge base is your team's job, not a change request to us.

At handover

The pilot

From your documentation
to a working assistant.

Start

Brief and a call

We sign an NDA and talk through the queue the chatbot will face and the documentation it must answer from.

Once the brief is complete

Audit and itemised proposal

We audit your documentation and confirm the date your proposal arrives: the questions in bounds, the sources, the escalation rules and the price.

Your decision

You sign off

You confirm the scope, the payment and the access we need. Nothing starts before this.

Typically 2 to 3 business days later

The team starts

We agree in writing what a correct answer and a mandatory escalation look like, and build the pilot on your own documentation.

On the date in your proposal

First deliverable

A working pilot that answers, cites sources and escalates on your rules. It then runs against real tickets, the helpdesk integration goes live, and it meets customers.

Inside Prism

Your engagement, week to week,
in one workspace.

Prism is the client workspace where your requests, approvals, tasks and weekly reviews live.

Where the engagement includes it, your AI build runs in a Prism workspace you log into: requests, approvals, the task list and the weekly review, with each decision and its expected result written down.

See how Prism runs an engagement

01
Requests and approvals
One queue, one owner, one due date.
02
Weekly review
Each decision recorded with its expected result.
03
Actions checked against outcome
What we did and what happened, side by side.

How we report it, in Prism

The eval scorecard,
before the pilot goes near a customer.

What you receive in month one

AI work is reported against an evaluation set built before any model is chosen. In the first month you receive the task definition with the inputs and acceptable outputs written down, the eval set of real cases with expected answers, and the scorecard showing accuracy, refusal behaviour and cost per run. The go/no-go checklist names what must hold before the pilot touches live traffic.

01Task definition: inputs, acceptable outputs, and the cases the system must decline
02Eval set: real cases with expected answers, reviewed by your team
03Scorecard: accuracy, refusal behaviour, latency and cost per run
04Go/no-go checklist: the thresholds the pilot must meet before launch
05Failure log: each wrong answer, its cause, and the fix applied

Related workB2B SaaS: from invisible to answer-engine cited.SaaS · UK · A different discipline, so it is linked here rather than presented as proof of this one.

The rest of the AI bench

A chatbot is one door in.

The retrieval layer under a chatbot is a knowledge system. The escalation logic is workflow automation. Start wherever the pain is, or start at the hub. To add chatbot developers to your own team instead, hire them by the month.

Knowledge / RAG Hire chatbot developers Workflow Automation All AI solutions

Who is accountable

The people accountable
for the work.

Leadership and how we’re built →

Parvessh Agarwal

Founder & CEO

An entrepreneur at heart who connects business analysis, strategy, clear objectives and ROI with practical technology solutions.

LinkedIn →

Saundaraya Gupta

Technical Manager

Brings technical delivery, client communication and project management together so work moves from brief to release smoothly.

LinkedIn →

Every engagement

A named coordinator and escalation contact

Know who coordinates your engagement, who owns the work and how to escalate an issue.

All engagements. Roles may be combined and vary with your engagement.

Full terms →

Before you sign

Our commitments

Each one says where it applies and links to its full terms.

Response and start times

We typically respond within four business hours. Eligible engagements can typically start within two to three business days after scope, payment and required access are confirmed.

Business hours are Monday to Friday, 9:00 am to 6:00 pm IST, excluding published company holidays. Typical expectations, not guaranteed service levels.

Full terms

Clear IP & Ownership Terms

Your materials remain yours. Bespoke deliverable rights, licences and handover are agreed before work starts.

All engagements.

Full terms

Clear Delivery & Escalation Ownership

Know who coordinates your engagement, who owns the work and how to escalate an issue.

All engagements. Roles may be combined and vary with your engagement.

Full terms

How we handle claims, credentials and your data: group credentials, dated ratings, how case figures are signed off, client permissions and data handling.

The doc audit

What we check first
on a chatbot engagement.

Before a pilot is scoped, we audit the queue the assistant will face and the documentation it must answer from. These are the checks, in the order we run them. Each one ends up written into the scope.

01The questions your queue actually asksWe map the queue the chatbot will face, so the pilot is built around the questions your customers send rather than the ones a demo would pick.
02Whether your documents answer themWe audit the help centre, policies, manuals and past tickets it must answer from. The rule is no source, no answer, so a question your documents do not cover goes to a person.
03Which tier each question sits inA question answerable from your documents needs an index. One that needs the customer's own record needs identity checks first. One that changes something needs write access, a confirmation step and an audit trail.
04What must always reach a personRefunds outside policy, legal questions, angry customers and anything ambiguous. We agree in writing what a mandatory escalation looks like, and the conversation and document that travel with it.
05What the evaluation will testThe suite is built from your real tickets: grounding, tone, attempts to talk it out of its rules, and the topics it must refuse. You see the scores and sign off before launch.
06Where it will liveThe helpdesk, CRM, website and channels you already run. The model underneath is chosen on fit, cost and data residency, and structured to be swappable.

Each tier costs something different to serve, so the pilot is scoped to one queue and the tier its questions sit in.

Questions

Frequently
asked.

It depends on the queue it faces and the systems it touches, so every engagement opens with a scoped pilot: one queue, your own documentation, a working assistant. The proposal itemises the questions in scope, the integrations and the guardrails and evaluation included. More channels or queues are priced only after the pilot has earned it.

It can, and no vendor can promise zero errors. Our job is to make them rare, visible and safe: the assistant answers only from your documentation, shows the source, and hands over to a human when confidence is low. An evaluation suite built from your real tickets runs before launch, and every conversation is logged afterwards.

No. A grounded chatbot clears the repetitive layer of the queue: the questions your documentation already answers, asked hundreds of times. Everything that needs judgement, such as exceptions, complaints and negotiations, goes back to your team with full context. What changes is where human attention goes, and we won't put an invented percentage on it.

It grounds in your help centre, policies, manuals and past tickets. Your data answers your customers and nothing else: we work under NDA, on enterprise API terms where providers don't train on your data, and inside your own cloud if needed. Retention, residency, access, deliverable rights, licences and handover are agreed before work starts.

The ones you already run. Builds integrate with your helpdesk, CRM, website and the messaging channels your customers use, and the model underneath is chosen on fit, cost and data residency, structured to be swapped without rework. If an off-the-shelf tool would serve you better than a custom chatbot, discovery ends with us saying so in writing.

Yes, with the same grounding discipline. A sales assistant answers product and pricing questions from your documentation, qualifies visitors with questions you approve, and books meetings or routes hot leads to a human. It won't improvise discounts, invent competitor comparisons or promise features that don't exist, and the guardrails cover sales scripts as strictly as support.

The range

What it costs,
at the direct rate.

Priced as a build on one queue, grounded in your own documentation, and a monthly support retainer that keeps it grounded as the documentation changes.

ProjectChatbot build$625–$10Kfixed quote, one-time
RetainerMonthly tuning and hosting support$250–$1,250per month

Indicative: every figure is refined into an itemised proposal.INR figures shown excl. GST.

Cost guideHow much does an AI chatbot cost in 2026?The options, the drivers and the range →Rate cardAI chatbots beside every other serviceDirect rate, in ₹ and $ →

Practitioner notes
on this work.

All articles

29 August 2026 · 5 min read

AI agent vs automation vs a chatbot: the questions we ask before building an agent

A chatbot answers, an automation follows a path drawn in advance, and an agent chooses its own next step.

25 August 2026 · 5 min read

Why governed delivery says no to some AI automation, and where the approval step goes

Most AI automation requests are good ideas. Some tasks should not run unsupervised however capable the model is. How governed delivery tells the two apart.

Put a grounded chatbot
on your busiest queue.

Scope a pilot: one queue, your own documentation, a working assistant with sources shown, escalation built in, and guardrails tested before a customer ever sees it. Tell us which tier your questions sit in; the pilot is scoped to that tier.

Get a Proposal White-label AI for agencies

Reply within four business hours (typical) · NDA standard · Clear IP & Ownership Terms

Last updated 21 October 2026

A chatbot that answersfrom your own documentation.

Judge the work,not the adjectives.

Every demo answers tier one.You are buying tier three.

Answerable from your documents

Needs this customer’s own record

Changes something

Judgement, exception or an upset customer

What a chatbot buildactually includes.

Grounded in your documentation

Escalation to a human, by design

Guardrails & evaluation before launch

On your stack, in your channels

Handover, logs and ownership

From your documentationto a working assistant.

Brief and a call

Audit and itemised proposal

You sign off

The team starts

First deliverable

Your engagement, week to week,in one workspace.

Requests and approvals

Weekly review

Actions checked against outcome

The eval scorecard,before the pilot goes near a customer.

A chatbot is one door in.

The people accountablefor the work.

Parvessh Agarwal

Saundaraya Gupta

A named coordinator and escalation contact

Our commitments

Response and start times

Clear IP & Ownership Terms

Clear Delivery & Escalation Ownership

How do we know the work will be good?

Who owns the ad accounts, the data and the work?

What if we want to stop?

How will we know what you are doing?

What we check firston a chatbot engagement.

Frequentlyasked.

What it costs,at the direct rate.

Practitioner noteson this work.

AI agent vs automation vs a chatbot: the questions we ask before building an agent

Why governed delivery says no to some AI automation, and where the approval step goes

Put a grounded chatboton your busiest queue.

A chatbot that answers
from your own documentation.

Judge the work,
not the adjectives.

Every demo answers tier one.
You are buying tier three.

What a chatbot build
actually includes.

From your documentation
to a working assistant.

Your engagement, week to week,
in one workspace.

The eval scorecard,
before the pilot goes near a customer.

The people accountable
for the work.

What we check first
on a chatbot engagement.

Frequently
asked.

What it costs,
at the direct rate.

Practitioner notes
on this work.

Put a grounded chatbot
on your busiest queue.