Skip to content
Sunbeam
Open menu

Services for businesses in Lucknow, India and beyond

Starter pilots are fixed-scope projects for one workflow, so you can see it work before you widen it.

A starter pilot is a small, fixed-scope project around one workflow. You see it working with your own staff and your own examples, inside a written line that says what it will not do. Pilots exist so a business can learn whether a system is worth widening, before commissioning the wider platform.

Sunbeam runs these from Lucknow with teams across India and with clients in the Gulf, the UK and the US. The investment is stated in a short proposal after we confirm the workflow. This page does not list prices. Each pilot below names the problem, what is included, what waits, and a typical time range.

The business case

What better looks like.

  • Working software on your own examples
  • A written edge around what the pilot will not do
  • A decision to widen, keep, or stop

What it is

One workflow, a written edge, and a decision at the end

A pilot is finished when the agreed path works for the agreed users: one location, one team, one collection or one channel. You leave with working software, a short handover, and a decision to widen, adjust or stop.

Who it is for

Teams that want evidence before a larger system

Pilots fit owners who can name one daily pain and show real examples in the same fortnight. If you already need several teams and a long integration list, start with discovery and use a pilot as the first slice.

  • One group of users

    One front desk, one sales seat, one venue or one internal team. A wider rollout is a decision after they have used it.

  • An observable check

    A booking is confirmed, a missing document is chased, a lead has an owner, or a question cites a source. That check is written into the scope.

  • Real examples

    A menu, a spreadsheet, a handful of documents, or the kinds of calls you actually receive. Invented samples teach the wrong workflow.

How we work

The edge is part of the scope

The proposal states the included path, the exclusions, the time range and the investment. Work starts when that is accepted. A second workflow is a new scope. You own the code and the data unless the proposal says otherwise.

  • Included

    The one path, the roles named in the proposal, a basic history of what happened, deployment you can access, and a short handover.

  • Left for later

    Extra sites, extra integrations, reporting beyond the agreed view, and any action marked as needing a person.

  • Your side

    Examples, access to the tools we must touch, and a reviewer. A pilot that waits a month on content has stopped being a pilot.

The pilots

Eight places a fixed scope is usually enough

Start with the workflow that happens every day. The linked service explains how the same idea grows after the pilot has proved itself.

  • Often 3 to 5 weeks

    Booking and reminders

    A request, a confirmation and a reminder on a channel you already use, for one team and one diary. Staff see the list and each booking's status. A later project can add multi-location calendars, payments and a full practice suite. Choose it when no-shows and a diary kept in two places are the pain.

    Business systems, if the diary should become the record
  • Often 3 to 6 weeks

    Client intake and document tracker

    One intake type: the form, the required documents, what has arrived, and a reminder for what has not. Staff see status without searching email. A later project can add case management, extra signature tools and retention rules your advisers still need to set.

    Business systems and document control
  • Often 2 to 5 weeks

    Lead follow-up automation

    An enquiry from one source gets an owner, a status and a timed follow-up, for one team. A later project can add a full CRM, scoring and campaigns. Choose it when enquiries arrive and then go quiet. A custom CRM is a later question, and only if configuration will not do.

    Automation and AI
  • Often 3 to 6 weeks

    Spreadsheet-to-dashboard

    One spreadsheet or export becomes a dashboard for a few agreed decisions: what is late, what crossed a threshold, what changed. Definitions come before charts. A later project can add more sources. Choose it when someone rebuilds the sheet before every meeting.

    Business dashboard development
  • Often 4 to 7 weeks

    QR table ordering pilot

    One venue, a menu, a table QR code and an order on a staff screen. Guests do not need an account. A later project can add card payments, stock across branches and a point-of-sale replacement. Choose it when one dining room still takes orders on paper.

    Websites and digital products
  • Often 3 to 6 weeks

    Document Q&A pilot

    One team, one approved collection, answers with citations, and a flag when an answer looks wrong. The owner can vouch for every file. A later project can add permissions across the company and an archive of scans. Choose it when people hunt the same folder every week.

    Knowledge systems
  • Often 4 to 8 weeks

    AI receptionist pilot

    One number or one channel, a written list of intents, your hours, and a hand-off to a person for everything else. Early conversations are reviewed before you rely on it. Outbound calling, professional advice, and unconfirmed changes to a booking or a payment stay out. Disclosure and recording follow the rules where you operate.

    Automation and AI, including reception
  • Often 3 to 8 weeks

    AI agent for one workflow

    One process, a short tool list, and a stop when the task leaves that list. A person approves messages to customers, money movement, and writes to a system of record. A second workflow is a later project. Choose it when the steps branch and you can show real examples.

    How an agent is bounded

Typical stack

The same foundations, in a smaller slice

A successful pilot should be able to grow, so it uses the same family of tools as our larger systems. Next.js is the screen, Postgres holds the records, Go runs a reminder or ingestion job when that work should continue on its own, and AWS is for files and a hosting account you should own. A very small pilot may be Next.js and Postgres alone, or an integration with a system you already have.

  • Interface

    Next.js

    The page a guest, a client or a staff member uses, including on a phone.

  • Services

    Go

    Reminders, webhooks and ingestion when that work should keep running with the admin screen closed.

  • Records

    Postgres

    The pilot's records and a simple history of what changed, enough to trust and later extend.

  • Operations

    AWS

    Hosting and private files when the pilot needs them, in an account you can keep.

Timeline ranges

Most pilots land between two and eight weeks

Each range starts once we have examples and a reviewer. Two weeks fits a narrow automation with access already granted. Eight weeks is the long end of a receptionist or an agent, usually because a provider or a consent decision takes another round. If the edge cannot be held, we propose a different shape.

  • A few days

    Confirming the edge

    Conversations and a proposal you can accept or decline: included work, excluded work, time range and investment.

  • 2 to 8 weeks

    Building the slice

    The clock runs while we have access and examples. Waiting on a menu, an export or a phone account pauses it.

  • The next fortnight

    Using it

    Live use is part of the pilot. We fix friction inside the original edge and list new ideas for a later decision.

Process

How a starter pilot runs

The steps are short so you always know whether you are still inside the agreed edge.

  • 01

    Name the workflow

    One paragraph in your own words, plus what a good week would look like.

  • 02

    Write the edge

    Included path, exclusions, the examples you will provide, the time range and the investment.

  • 03

    Build the path

    You see it on your examples and tell us where a real case would fall off.

  • 04

    Hand it to the pilot group

    The people named in the proposal use it. We watch the success check.

  • 05

    Hand over

    Access, a short operating note, and the list of ideas that were left out on purpose.

  • 06

    Decide

    Widen into the related service, run a second pilot, or keep this slice as the whole system.

Who this is for

For teams ready to improve an important part of the business.

  • Owners who can name one daily pain
  • Teams with real examples ready to share
  • Sponsors who want evidence before a larger system

Frequently asked questions

Useful questions before you begin.

Are starter pilots fixed scope?

Yes. The proposal states the path, the exclusions and the investment. A second workflow is a new proposal.

What happens if we want to continue?

The pilot can widen into the related service when the foundation is sound. You can also keep the slice as it is, or stop.

Why are there no prices on this page?

The figure depends on the workflow, the integrations and how ready your examples are. It is stated in the proposal after a short confirmation.

How do we choose a pilot?

Pick the workflow that happens every day and that you can show with real examples. If two qualify, start with the one that already costs you time each morning.

Related capabilities

Considering whether a custom approach is justified? Read our build-versus-buy decision guides, explore industry workflows, or review how ownership and handover work.

Start with the business problem, not the technology.

Tell us what is getting in the way. We will help identify a useful next step.