Bespoke software

I want to use AI. Where do I start?

No jargon and no assumptions. An honest conversation about what AI can do for your business, and then we build it.

The honest answer

If you have said that out loud in the last year, you are not alone. Most business owners know AI matters. Very few know what to do about it.

The problem is not a lack of AI tools. There are hundreds of them. Most AI advice assumes you already have clean data, connected systems and a clear use for it. Most small businesses do not have those yet, and almost nobody talks about that part.

We start where you are

Before we recommend anything, we want to understand your business. What data do you have? Where does it live? How clean is it? What are you trying to achieve? The answers change everything. It might be a simple connection to a tool you already use, or a local AI model on your own infrastructure. We have no one-size answer, because there isn't one.

  1. 1Work out what you have got

    Most businesses are sitting on more useful data than they realise. The trouble is that it is scattered across systems, formatted inconsistently or not connected. Before AI can do anything useful, you need to know what you are working with.

  2. 2Make sure it is clean

    Messy data produces useless AI output. Getting your data structured, connected and reliable is not glamorous work, but everything else sits on it.

  3. 3Work out what to do with it

    Once the data is in good shape, the options open up: automated summaries, alerts when something needs attention, patterns across your pipeline or operations, and local AI that never leaves your building. The right answer depends on your business.

AI we have already built

  • A plain English AI summary of sales and operations that IceBubble CRM users read every morning
  • An AI survey builder that turns a description into a working form, with questions, input types and rules
  • An API and MCP connection that lets assistants such as Claude read and update a CRM within set permissions
  • Integrations between systems that were never designed to talk to each other, which is where most AI projects really start

The AI layer is new. The data work underneath it is what we have done for over a decade.

What it looks like in practice

It starts with a free 30 minute discovery call. No pitch and no proposal, just an honest conversation about where you are and whether we can help.

If it makes sense to go further, we scope what needs doing and give you a clear picture of the options, the cost and what you would get at the end. You decide whether to go ahead.

Frequently asked questions

We do not know where to start. Is that a problem?

No, that is where most people are. The first conversation is about your business and your data, not about AI.

Will our data be sent to an AI company?

Not unless that is the right answer for you. Depending on what you need, AI can run through a service such as Claude or OpenAI, or as a local model on your own infrastructure, keeping data private and GDPR safe.

What if we are not ready?

We will tell you, and tell you what to sort out first. We would rather have that conversation now than start a project that is not right for either of us.

What does it cost?

It starts with a free 30 minute discovery call. If it makes sense to go further, we scope what needs doing and give you a clear picture of the options and the cost before you decide.

Other things we do

Web applications

Systems built around how your team works, from ordering platforms to client portals.

More on web applications

App development

Apps for staff and customers on iPhone, Android, Windows and Mac, from one codebase.

More on app development

Book a free 30 minute discovery call

Tell us what's slowing your team down and we'll tell you honestly whether IceBubble CRM, a bespoke build or something off the shelf is the right answer. No hard sell.

Book a call