The best CRM options for small teams (including not buying one)

The CRM market is enormous, mature and mostly aimed at sales organisations considerably larger than yours. That makes choosing hard, because the loudest options are optimised for a buyer you are not.

Start with the question nobody selling a CRM will ask you.

On pricing: CRM pricing and tier structures change constantly, and free tiers get restructured. This article deliberately does not quote prices or limits. Verify directly, and pay attention to which features sit on which tier — the useful ones are often a tier above where you expect.

First: do you need a CRM platform at all?

Three honest categories.

You need a full CRM platform if you have a sales team with pipeline stages people disagree about, quota and territory management, sequences and cadences, or a marketing function that needs to hand leads over formally. Buy a real one; assembling this yourself is a bad use of your time.

You need a lightweight CRM if a few people sell, deals take weeks not hours, and you mainly need to remember who to follow up with. Most of the market fits here.

You need a relational workspace, not a CRM if your sales process is unusual, or — more commonly — if the customer record needs to connect to things a CRM does not model: your inventory, your projects, your deliveries, your support history. This is the case CRMs handle worst, because they own the customer record and refuse to share it.

That third category is more common than the market admits. If your real problem is "the customer record needs to link to our operational data", a CRM makes it worse by creating a second source of truth.

The lightweight options

Broad strokes, because the details change:

Pipedrive — built around the pipeline, and unapologetic about it. Simple, visual, quick to adopt. Weakest if you need it to do much beyond deal progression.

HubSpot — a free tier that is genuinely usable, and a very large product above it. The strength is the breadth; the risk is that the features you eventually want sit several tiers up, and the price step is steep. Fine if you go in knowing that.

Zoho CRM — extensive, configurable, part of a large suite. Good value; the interface is denser than the others and it takes more setting up.

Folk, Attio and similar newer tools — relationship-first rather than pipeline-first, often better for consulting, agencies, investing and partnerships where the network matters more than the funnel. Worth a look if "pipeline" has never quite fitted how you work.

Close, Copper and similar — vertical or workflow-specific bets. Copper leans heavily on Google Workspace integration; Close leans on high-volume calling. If their specific bet matches how you sell, they are strong.

How to choose between them: pick the one whose default view matches how you actually work. Every one of these can be configured into the others. The one that fits without configuration is the one people will still be using in six months.

The build-it-yourself option

If you land in the third category, building a CRM in a relational workspace is genuinely viable — and the reason is not cost. It is that the customer record can link to everything else.

The model is four tables:

• Companies — the organisations.
• People — the humans, separate, because people change jobs and you want to keep their history when they do.
• Deals — the opportunities.
• Activities — every meaningful touch.

Activities is the table that makes it work. From it you get last-contact date, touches-to-close, and a neglected-accounts view — all as rollups rather than reports.

When building beats buying:

• Your customer record must link to inventory, projects, deliveries or support tickets that already live somewhere.
• Your process genuinely does not match a pipeline.
• You need field-level access control (contractors see accounts, not margins).
• You are already running operations in a relational workspace and a separate CRM would create a second source of truth.

When buying beats building:

• You want email sequences, call logging and meeting scheduling out of the box. Rebuilding these is not worth it.
• Your process is a normal pipeline. Do not rebuild a solved problem.
• Nobody on the team wants to own a build.
• You need a mature mobile app.

That first "buying" point deserves weight. Email integration — logging conversations against records automatically — is the single most valuable CRM feature and the most annoying to build. If you need it, buy.

The spreadsheet answer

For a genuinely small operation — one or two people selling, twenty active deals — a spreadsheet is fine and everyone pretending otherwise is selling something.

The line is where it stops being fine:

• More than one person edits, and you have started coordinating who touches it.
• You have lost a deal because nobody followed up.
• The same company appears on several tabs, spelled differently.
• You cannot answer "when did we last speak to them?"

That last one is the real signal. A CRM's core value is not the pipeline chart. It is knowing who has gone quiet, and a spreadsheet cannot tell you without an activity history — which is exactly the thing spreadsheets are worst at.

What actually determines success

Whatever you choose, the deciding factor is not the tool.

A CRM that is 80% modelled and current beats one that is 100% modelled and three months stale. Every implementation that failed, failed because updating it was work that benefited someone else.

So:

• Require three fields, not fifteen. Company, value, next step.
• Make "next step" mandatory and short. It is the only field that helps the person filling it in, which is why it is the one that keeps the CRM alive.
• Define stages by observable events, not feelings. "Proposal sent" is verifiable; "Interested" is not, and forecasts built on it are fiction.
• Build the salesperson's daily view before the founder's dashboard. If the doer's view is not the fastest way to answer "what am I doing today?", the data behind the dashboard will be stale.

The recommendation

Sales team, standard pipeline — buy a lightweight CRM. Pick the one whose default view matches your process. Do not over-configure it.

Customer data needs to connect to operations — build it in a relational workspace, or you will maintain two customer lists forever.

One or two people, twenty deals — a spreadsheet, until you miss a follow-up. Then revisit.

Sales organisation with quota, territory and marketing handoff — a full platform, and this article is not aimed at you.

The most expensive outcome is buying a platform sized for a company three times yours, spending a quarter configuring it, and having it go stale because nobody wanted to fill in fourteen fields per deal. That happens far more often than choosing a slightly suboptimal tool.

Next: how to build a CRM you'll actually use.