Most CRM projects fail on adoption, not features. Choose for the people who will use it every day.

How to choose a CRM for a growing UK business

A practical way to shortlist and choose a CRM - what to decide before you look, the questions that separate suppliers, and the traps to avoid.

There are hundreds of CRM products, and most of them will demonstrate beautifully. Almost every one can store contacts, show a pipeline and send an email. The differences that decide whether a CRM works for you are rarely visible in a demo.

This guide sets out how to make the decision in a way that holds up six months later: what to decide internally first, what to look for, what to ask, and what to be wary of.

Decide three things before you look at anything

The most useful work happens before the first demonstration.

  1. 1The problem you are solving, in one sentence. "Enquiries go unanswered" or "nobody trusts the forecast" leads to a different choice from "we need a CRM".
  2. 2Who will use it every day, and what they do now. The system has to suit the salesperson in the car and the person on the phones as well as the manager reading reports.
  3. 3What must connect to it - email, accounts package, website forms, phone system - and which of those are non-negotiable.

What actually matters in daily use

The features that decide adoption are unglamorous: how quickly you can find a customer, whether email files itself, how few clicks it takes to log a call, and whether the next step on each deal is obvious.

  • Email and calendar sync that works without anybody pressing a button.
  • A customer record that shows everything - contacts, conversations, quotes, cases - on one screen.
  • A pipeline the team can update with a drag, on a phone as well as a desktop.
  • Quotes built from your price list, with follow-ups scheduled automatically.
  • Permissions that let you share what should be shared and restrict what should not.
  • Reporting that answers your top five questions without an export.

Look past the headline price

CRM pricing varies enormously in shape. Some products are priced per user with everything included; others split features into tiers, charge separately for marketing contacts, sell add-ons for quoting or telephony, or require an onboarding fee.

The only fair comparison is the total cost of the configuration you actually need, for the number of users you expect in two years, over three years. Ask every supplier for that figure in writing.

Be wary of choosing on a starter tier. Check which tier contains the features you are buying the system for, and price that one.

Questions that separate suppliers

  • Which of our requirements are standard, which need configuration, and which would need custom work?
  • Who will set it up, and who supports it afterwards? Are they the same people?
  • Where is our data hosted, and how do we get all of it out if we leave?
  • How does email sync handle personal and internal email?
  • What does it cost in year three with the users and features we expect then?
  • When would you tell us we are the wrong customer for you?

Common traps

  • Buying for the most complex process you might have one day, and making the everyday work harder.
  • Letting the demonstration drive the requirements rather than the other way round.
  • Choosing a platform that needs a specialist administrator you do not have.
  • Assuming data import is trivial. It is where most of the effort goes.
  • Forgetting about growth - what happens when you also need orders, stock or operations?

In short

What to take away

  • Decide the problem, the users and the must-have connections before looking.

  • Choose for daily use: email filing, one-screen records, quick updates.

  • Compare the three-year cost of the configuration you actually need.

  • Ask every supplier when they would tell you they are the wrong fit.

Questions

Asked while people are working this out

  • Not necessarily. The market leaders are excellent for large organisations with administrators to run them. For a small or mid-sized team, a simpler system that everybody uses will almost always outperform a powerful one that nobody updates.

  • Three is usually enough. More than that and the demonstrations blur together. Use your problem statement to rule out the rest quickly.

  • A trial with sample data tells you about the interface. A walkthrough using your own customers, a real enquiry and a real quote tells you whether it fits. Ideally, do both.

  • It depends on your customers and contracts. UK hosting can simplify data protection documentation and tender questionnaires, and some customers ask for it. Check whether yours do.

After reading

Bring the questions this raised

If the guide has made you think of something specific about your own business, that is exactly the conversation worth having. No obligation and no scripted demonstration.