Astraeus

CRM and enquiry systems

A CRM the team actually uses, from selection to support

Astraeus consults on, configures, deploys and supports CRM systems for SMEs. The system arrives working: records, pipelines, automations and the AI tools your team already pays for connected around it.

Book the consultation
Marble figure guiding scattered enquiry paths into one ordered CRM pipeline

Most CRM projects fail after the invoice, not before it. The software gets bought, half-configured and quietly abandoned, because nobody shaped it around how the team sells, migrated the records that matter or trained the people who touch it daily. The gap is not software. It is deployment.

That deployment is the service: choose the platform, configure pipelines and permissions around the real sales process, migrate the data worth keeping, train the team and stay available after go-live. We run our own pipeline on HubSpot and Attio daily and build against their APIs and AI integrations, so the advice comes from use, not from a feature list.

Once the CRM is the system of record, it becomes the hub everything else plugs into: enquiry automation, meeting capture, reporting and the AI assistants the team already uses. That layer is the second half of this page.

What this can look like in an SME

The tools are examples, not a fixed stack. Access, cost, security and the way the team already works decide the final design.

Trades and home services

Quote requests land by phone and website form. During a busy fortnight the calls get answered and the forms wait, and nobody knows which quotes were never sent.

What we could implement

A HubSpot pipeline shaped around enquiry, quote and job. Every form and logged call creates one record with an owner, an acknowledgement goes out with a realistic timeframe and each open quote gets a follow-up task at three days.

Tools often involved
HubSpot free tier, website forms, a simple mobile habit for call notes
Human checkpoint
The quote itself and any pricing conversation.

Professional services

Client calls carry the real detail: scope, objections, commitments. None of it reaches a system, so follow-ups are rebuilt from memory and referral sources are guesswork.

What we could implement

Fireflies records and summarises the call, the summary and action points land on the right CRM record automatically, drafted follow-ups wait for approval and a monthly view finally shows conversion by referral source.

Tools often involved
Attio or HubSpot, Fireflies, Gmail or Outlook
Human checkpoint
Every client-facing reply until the drafts prove themselves.

A sales team already using Claude or ChatGPT

Reps research prospects and draft proposals in AI tools all day, but none of that work touches the CRM, so the pipeline is always a week out of date.

What we could implement

An MCP connection gives Claude or ChatGPT controlled access to the CRM. Reps pull pipeline context, draft follow-ups from the actual record and log outcomes without leaving the tool, and scoped permissions decide what AI can read and write.

Tools often involved
HubSpot or Attio, Claude or ChatGPT, an MCP server scoped to the CRM
Human checkpoint
Anything sent to a customer and any change to deal value or stage.

The sequence

How the work moves

  1. 01

    Consult

    Map how the team sells today, what must be recorded and which platform fits: HubSpot, Attio, Zoho, Pipedrive or the one already paid for.

  2. 02

    Configure

    Set up pipelines, stages, fields, permissions and views around the actual sales process, not the vendor default.

  3. 03

    Migrate and train

    Bring across the records worth keeping, connect forms, inbox and calendar, and train the team on the handful of habits the system depends on.

  4. 04

    Support

    Stay available after go-live: an adoption review after a month of real use, then automations and integrations as the business asks for them.

What a deployment covers

Platform selection

A recommendation against how the team sells, the budget and what is already paid for, with the reasoning written down.

Configuration

Pipelines, stages, custom fields, views and dashboards shaped around the real process, so the CRM describes the business rather than the vendor demo.

Data migration

Spreadsheets, inboxes and old systems mined for the records worth keeping, cleaned and imported with sources intact.

Permissions and structure

Who sees what, who can change what and what gets logged. Decided before go-live, not after the first mistake.

Training and adoption

Sessions built around the team's own records, plus the two or three daily habits the system depends on.

Documentation and support

Operating notes, a named contact and a review after a month of real use. The CRM account and the data are yours from day one.

Automations that work for people

Automations that earn the CRM its keep

A CRM full of stale records is a filing cabinet. These keep it current and make it pay.

Small-business owner and colleague reviewing a customer pipeline together on a laptop
Illustrative CRM working session
Before

Forms, inboxes and call notes sit in separate places.

Managed

1. Enquiry capture

Every enquiry creates one record with a source and named owner.

Human handover

The owner picks up the conversation.

Before

Conversations stall because nobody knows what needs chasing.

Managed

2. Follow-up that stops

Tasks and reminders end the moment the customer replies.

Human handover

A person handles every live reply.

Before

Call detail and commitments are rebuilt from memory.

Managed

3. Meeting capture

Summaries, actions and commitments land on the right record.

Human handover

The team confirms what was agreed.

Before

Open quotes and stale deals disappear into the week.

Managed

4. Quote and pipeline chasing

Due work surfaces with its owner and the next action visible.

Human handover

Commercial decisions stay with the owner.

Before

The Monday picture is rebuilt by hand from several sources.

Managed

5. Reporting

Enquiries, response time, quotes and wins stay visible by source.

Human handover

People interpret the numbers and decide.

Automations do the capturing and chasing. People bring the judgement.

The AI layer

The CRM becomes the system your AI tools plug into. Access is scoped and permissions are designed before anything connects.

Claude and ChatGPT connections

Is the team already on Claude or ChatGPT? An MCP connection lets those tools read and update CRM records with scoped permissions: pull pipeline context, draft follow-ups from the live record, log the outcome.

Custom skills

AI assistants can be taught the playbook: how the business qualifies, what a good follow-up sounds like, which fields matter. We build these for our own Attio pipeline and can build them for yours.

Call and meeting intelligence

Transcripts become searchable context on the record: what the customer said, the objections raised, the commitments made.

Guardrails

Each connection gets only the access its job needs. Money, customer commitments and sensitive changes stop for a person.

Are you getting the most from your integrations?

Most teams pay for tools that never talk to each other. These are combinations we configure, connect and support.

HubSpot · Fireflies · Gmail

Calls recorded and summarised onto the record, email threads tracked automatically, follow-ups drafted for approval. The rep talks; the CRM keeps itself current.

Attio · Claude · MCP

The pipeline becomes something you can talk to: ask which deals went quiet, draft the chase from the record, log the outcome. Scoped permissions decide what AI touches.

HubSpot · website forms · WhatsApp

Every route a local customer actually uses lands in one queue with an owner, including the enquiry that arrives at 9pm.

CRM · calendar · booking

A booking link in the first reply removes two rounds of email. The meeting, the record and the follow-up task connect themselves.

Native, middleware or custom

Connections use the built-in integration where it covers the job, Make, n8n or Zapier where routing is needed, and a custom API or MCP build where nothing off the shelf fits.

Choosing the CRM

HubSpot

The usual default for an SME starting out. The free tier covers capture, pipeline and email, and there is room to grow into marketing and service work without changing systems.

Attio

Fast, flexible and data-first. Suits teams who want the CRM shaped around relationships and their own way of working rather than a fixed pipeline.

Zoho and Pipedrive

Worth assessing when cost, an existing subscription or a specific feature points that way.

The one you already have

A CRM the team ignores is usually a process problem, not a purchase problem. We would rather fix the habits than sell a migration.

The honest boundary

We use HubSpot and Attio in our own business daily, which informs the advice and also biases it towards tools we know deeply; where Zoho, Pipedrive or the system you already pay for fits better, we will say so. And no CRM survives a team that will not record its work, so part of every deployment is agreeing the smallest set of habits the system depends on.

Direct answers

Common questions

Which CRM do you recommend?

It depends on the team. HubSpot suits most SMEs starting out because the free tier covers capture, pipeline and email. Attio suits teams who want the CRM shaped around relationships and data rather than a fixed pipeline. An existing CRM the team ignores is usually a process problem, not a purchase problem.

We bought a CRM before and nobody used it. Why would this be different?

Because adoption is designed rather than hoped for. The deployment connects the places work already happens, so records create themselves, and the team is left with two or three habits instead of a data-entry job. If the habits prove unrealistic, we change the design.

Can our AI tools connect to the CRM?

Usually, yes. HubSpot and Attio both expose APIs that Claude and ChatGPT can work with through an MCP connection, and we scope the permissions so AI reads and writes only what its job needs. The setup is tested against your real records before the team relies on it.

Will automated follow-up annoy customers?

Not if it is designed to stop. Every sequence ends the moment the customer replies, commercial commitments always come from a person and each message is written with the business in its own voice.

What does a setup cost?

Most first deployments land inside the €395 to €1,950 range quoted after a free consultation. Migration size, the number of connected channels and the AI layer are the main variables, and the quote comes before any commitment.

What happens to the data if we part ways?

It is your CRM account and your records from day one, with documentation handed over. Nothing in the design should depend on keeping the original builder.

Bring the pipeline, wherever it lives.

Spreadsheet, inbox or a half-used CRM: a free consultation maps how the team sells today and what a proper deployment would change.

Book the consultation