How to choose a CRM system

Most CRM selections begin in the wrong place, with a search for the best CRM and a set of demo bookings. The trouble is that every mainstream platform demos well, because the demo is built on clean data and a process designed to flatter the software.

The businesses that choose well start by writing down how they sell, market and support customers today, where that breaks, and what they need to be able to see. That document does two jobs: it cuts a long list to a shortlist quickly, and it makes vendor quotes comparable because everyone is pricing the same scope.

It also protects the project later. When a disagreement arises six weeks into implementation about whether something was in scope, a written requirement settles it in minutes.

Sales process. The stages a deal moves through, who owns it at each point, what has to be true to advance, and how quotes are produced and approved. If your process has genuine variants, for example new business against renewals, describe both.

Marketing. How leads arrive, what happens to them, what segmentation you need, and whether campaign attribution matters enough to pay for it. Be honest about the resource available to run marketing automation, because unstaffed automation is a licence cost with no return.

Service. Whether customer enquiries need to be logged as cases with response targets, whether you offer a portal or knowledge base, and how service history should be visible to the people selling to the same account.

Reporting and integration. The specific reports the business runs on, and the systems the CRM must connect to: email and calendar, your finance or ERP system, your website forms, and any sector-specific tool. Integrations discovered late are the most common cause of a budget moving.

Give every vendor the same script. Three or four scenarios from your own business, sent in advance, with a small sample of anonymised data. Ask to see an enquiry captured, qualified, quoted, won and handed to delivery, and ask to see the reporting that comes out of it.

Ask what you are looking at. Standard configuration, a setting, or custom development. That distinction matters more than anything else on the screen, because custom code is what makes future upgrades slow and expensive.

Have the people who will use it in the room. Salespeople notice friction that a project sponsor never will, and a system that adds ten seconds to every interaction will be abandoned no matter how good the dashboards look.

Score against your requirements immediately afterwards, while each demo is distinct. Left to the end of the process, everything blends into a general impression of the vendor who presented best.

Build one three-year total per vendor. Licensing by role, implementation services, data migration, each integration, training, and support in years two and three. Vendors present these differently, so normalise them into the same table yourself.

Watch for capabilities that appear in the demo but sit in a higher tier or a separate product: marketing automation, customer portals, advanced forecasting, telephony and embedded analytics are the usual ones. Price the edition you will actually need.

Include your own team's time. Requirements, data cleaning, testing and training all consume internal days, and projects that pretend otherwise slip. It is a real cost even though no one invoices for it.

Then judge value rather than price. The difference between the cheapest and the best-fitting option is often recovered in the first year through work that stops being manual, while a poor fit gets replaced within three years at the full cost of doing it twice.

Choosing on the demo. Every product looks capable when it is presented on clean data by the person who built the scenario. Judge fit against your written requirements instead.

Ignoring data quality. Duplicated contacts and dead companies carried into a new system destroy trust in it within weeks, and once people stop believing the data they go back to their own spreadsheets.

Configuring for management rather than users. If the only benefit of filling in the CRM accrues to the person reading the reports, it will be filled in badly. Give users something back on day one.

Selecting the partner last, or on price alone. The implementation team has more influence on the outcome than the platform. Ask who is actually doing the work, what else they have delivered in your sector, and who you will speak to after go-live.

If you are still deciding whether a CRM is needed at all, our guide to what a CRM system is covers the features, the comparison with spreadsheets and ERP, and the signals that a business has outgrown its current way of working.

If you are ready to evaluate specific platforms, Microsoft Dynamics 365 is the CRM we implement most often for UK organisations, and our pricing guide sets out per-user costs and what implementation typically involves.

And if you would like a second opinion on your requirements before you start booking demos, we are happy to look through them with you. It usually takes an hour and it tends to save weeks.