A guide to choosing a CRM without the marketing phrases - what to settle beforehand, the five questions to ask a supplier, how off-the-shelf and adaptable CRMs differ, and when the cheapest option is to buy nothing.
Choosing a CRM in a small company usually goes like this: somebody searches for “best CRM”, works through three websites full of the same words, picks the one with the nicest screenshots, and a year later finds the team uses three modules out of twelve while the things that matter are still in a spreadsheet.
This guide tries to break that pattern. It contains no league table. It contains the questions to ask yourself before you trial anything.
1. First: do you even need a CRM?
The honest opening chapter. A CRM solves one thing - visibility of relationships and sales. If your real pain is elsewhere, you'll buy a tool that won't fix it.
You need a CRM when: there are more opportunities than anyone can hold in their head, more than one person works on sales, enquiries get lost, or the history of client correspondence lives only in one person's mailbox.
You don't need a CRM when: you have five long-standing clients and no acquisition, or when your pain is in delivery and invoicing rather than in sales. In the second case you're looking for a job management system, not a CRM.
The most common case in a small company: you need both, and nobody tells you, because every supplier sells only one half.
2. What to settle beforehand
Have answers to these before you open the first demo. Without them, comparing tools is comparing screenshots.
Which entities you track. Contact, company, opportunity - every CRM has those. But do you also have a property, a vehicle, a case file, a project? If so, you're looking for an adaptable system, not a CRM.
What stages your sales process has. Write them down in your own words. If you have a site survey between “enquiry” and “quote”, you need to see it in the system - otherwise you'll never notice that deals get stuck there.
Which fields you keep on the side. Lead source, promised date, margin. This list is the single most important thing from the whole preparation: these are exactly the fields that will cause a spreadsheet to appear next to your new CRM.
Who must not see what. A commission-based salesperson and purchase prices. A temp and every client contact. A contractor and other people's jobs.
What the CRM has to connect to. Invoicing, email, your website, your online shop, accounting.
3. Off-the-shelf vs. adaptable CRM
The crucial difference, and one you can't read off a pricing page.
An off-the-shelf CRM is designed for a very broad audience - a sales team, a manufacturer, an agency and an estate agent at once. To suit everyone, it contains modules for everyone. What you need from it is the intersection, and that's usually small.
An adaptable CRM lets you set the fields, stages and rules to match how you actually work. The price of that is having to know how you work - which is homework nobody can do for you.
You bought twelve modules. You use three. And the spreadsheet is still there.
The unopened modules aren't the problem. The problem is that everything important the CRM doesn't cover gets tracked by each person their own way.
In detail: You're paying for a CRM you use 20% of.
4. Five questions for a supplier
These reveal more than an hour-long demo:
- Can I add a field myself? And can someone in the back office do it, or is it a request to you?
- Can I relate two databases that weren't originally connected? This separates genuinely adaptable systems from those that merely have configurable styling.
- How long does a change take and what does it cost? A system where adding a field takes three weeks and costs €600 trains you to stop asking.
- Can we get our data out? At any time, structured, at no extra charge.
- Where does the data physically sit and is there a processing agreement? A question worth asking before an incident rather than after one.
5. A test you can run in an hour
When you trial a specific tool, try these three things in the first hour. If you can do them yourself, you'll still be able to in two years:
- Add a custom field - say “lead source” with five options.
- Reorder the sales stages to match your process rather than the default template.
- Hide one column from one role - typically price from sales.
Three edits that mean a support ticket elsewhere
A new field, reordered stages and a hidden price column. If you can do them yourself, you know what every future change will look like.
In detail: What a CRM you can edit yourself looks like.
6. What it really costs
The subscription is only part of it. The budget has to include:
| Item | Where it hides |
|---|---|
| Licences per user | On the pricing page, visible |
| Implementation and setup | One-off, usually underestimated |
| Data migration | Depends on the state of your spreadsheets |
| Training | Grows with headcount |
| Changes over time | Nowhere - and it decides |
| Tools around it | Email marketing, invoicing, connectors |
The last two rows determine the three-year total far more than the first one.
In detail: What a custom app costs.
7. Specific comparisons
If you already have a shortlist, there are detailed comparisons of individual platforms on the site - data model, automation, email, printing and price side by side:
- Apexloop vs. Raynet - the most common choice among Czech sales teams
- Apexloop vs. Freelo - when projects are in scope too
- Apexloop vs. Caflou - a Czech tool with ready-made modules
- Apexloop vs. Notion - when you're considering building your own
- Apexloop vs. Salesforce and vs. Dynamics 365 - enterprise options
8. The most common mistakes
Choosing by feature count. The longest list wins and your team loses, because they can't find their way around it.
Skipping the question “which fields do we keep on the side”. The most reliable way to buy a system that will have a spreadsheet next to it within a month.
Deciding without the people who'll work in it. The owner picks on reporting, the team doesn't use it.
Underestimating migration. Data in a poor state doesn't improve on its own. Sometimes it's best to start with the last year and leave older records in an archive.
Buying when the pain is elsewhere. If jobs are getting lost in delivery, a CRM won't fix it.
Common questions about choosing a CRM
What should a CRM cost for a ten-person company?
The spread is wide, from tens to hundreds of euros a month depending on how many surrounding tools you need. More useful than hunting for an average is working out your own three-year total including changes and implementation.
Should we start simple and move to something bigger later?
Reasonable, with one condition: confirm you can get your data out of the simple one. Most painful migrations happen where you couldn't.
Is a local or an international tool better?
It depends how much you need local specifics - business registry lookups, invoice formats, support in your language. For purely sales work the origin of the tool is secondary; the moment invoicing and documents are involved, it starts to matter.
How long does a CRM rollout take?
Setup takes days, adoption takes weeks. A realistic plan assumes the team grumbles in month one, settles in month two, and comes back with change requests in month three - and that's exactly when you find out whether you chose well.
What if we need a CRM and job management?
Then you're looking for a platform with its own data model, not a CRM. Two separate tools create exactly the boundary where context gets lost - as described in Three systems that don't talk.