CRM Database Explained: Contacts, Companies, Deals and Activities

If you’ve just opened a CRM for the first time and stared at a screen full of tabs and record types wondering what any of it actually means, you’re not alone. A CRM database is simply a structured store of information about the people, organisations, opportunities, and interactions that make up your entire customer relationship history – organised so your team can act on it, not just read it.

That definition sounds simple. The tricky part is understanding why the database is split into these specific record types, and how they connect to each other in ways that actually change how you sell. Let’s work through each one.

What Is a Contact Record in a CRM Database?

A contact is a person. That’s it. In CRM terms, it’s a record representing an individual human being your business has some kind of relationship with – a prospect, a buyer, a decision-maker, or even someone who attended a webinar once and never replied to your follow-up email.

What makes a contact record genuinely useful isn’t the name and email address. Those are just identifiers. The value comes from everything attached to the record over time: which emails they opened, which calls were logged, what deals they’re associated with, and how recently anyone on your team actually spoke to them. A contact with no activity history is barely more useful than a business card in a drawer.

Typical fields you’ll find on a contact record include:

• Full name, job title, and email address
• Phone number and preferred contact method
• The company they work for (linked to a separate company record)
• Lead source – how they originally came into your database
• Owner – which team member is responsible for this relationship
• Custom fields your team adds based on your specific Ideal Customer Profile (ICP)

One thing worth understanding early: contacts and companies are separate record types that link to each other. A contact belongs to a company, but the company record has its own data. This distinction matters more than it first appears.

What Is a Company Record – and Why Keep It Separate?

A company record (sometimes called an Account, depending on which CRM you’re using) represents the organisation itself – not any single person within it. This is where you store firmographic data: industry, company size, annual revenue, website, and location.

Keeping companies separate from contacts solves a real problem. People leave jobs. A contact you’ve been nurturing for eight months might move to a different organisation next quarter. If contact and company data were merged into a single record, you’d lose the institutional history of your relationship with that business the moment your point of contact walks out the door. With a proper company record, that history stays attached to the account – the person simply gets a new employer linked to their contact record.

Company records also let you see the full picture of a business relationship at a glance: all the contacts who work there, all the deals you’ve run with them, and all the support interactions or notes your team has logged. For anyone thinking about Net Revenue Retention (NRR) or expansion revenue, this account-level view is where that analysis starts.

If you want a deeper look at how companies and contacts relate to each other structurally, our article What Is an Account in CRM? Companies, Contacts and Hierarchy covers the hierarchy in detail.

What Is a Deal Record in a CRM Database?

A deal (also called an opportunity in some platforms) is a record that represents a specific potential sale. It’s not a contact. It’s not a company. It’s the transaction itself – the thing you’re trying to close.

This distinction matters because one company might have three separate deals with you simultaneously, and a contact might be involved in two of them. Without a deal record that sits independently and links to both the contact and the company, your data becomes impossible to untangle.

A deal record typically includes:

• Deal name and value
• Current pipeline stage – where it sits in your sales pipeline
• Expected close date
• Associated contacts and company
• Deal owner
• Probability of closing, which feeds your sales forecast

The expected close date and deal value are what your manager is looking at when they pull a forecast report. Get those fields wrong – or leave them blank – and your pipeline data becomes meaningless. This is one of the most common places CRM adoption breaks down in practice: reps create deal records but don’t maintain them, so the data drifts away from reality within weeks.

Deal stages deserve a mention here. Most CRMs ship with defaults like “Prospecting,” “Qualified,” “Proposal Sent,” and “Closed Won.” These stages are customisable, and they should reflect how your team actually sells – not some generic template. The sales cycle at a company selling enterprise software looks nothing like the one at a business selling short-contract services, and your stages should reflect that difference.

What Are Activities – and Why Do They Matter More Than People Think?

Activities are the records of what actually happened. Calls, emails, meetings, notes, tasks – every interaction between your team and a contact gets logged as an activity and attached to the relevant contact, company, or deal record.

This is where a CRM database earns its keep. Without activity logging, you have a list of names and numbers. With it, you have a full history that any team member can read to understand the relationship before making contact – no more “wait, did we already pitch them on the enterprise plan?” moments.

Activities generally fall into a few types:

• Calls – logged manually or captured automatically via dialers integrated with your CRM
• Emails – often synced automatically if you connect your email client
• Meetings – calendar events linked to CRM records
• Notes – freeform text your reps write after an interaction
• Tasks – future actions someone on the team needs to take, with a due date and owner

Tasks are particularly underrated. A task attached to a deal record saying “Follow up after board meeting on Nov 12” is what stops a warm opportunity from going cold because it fell out of someone’s head. The CRM becomes the external memory the team actually relies on.

Activity data also feeds into some of the more advanced reporting you’ll encounter later – things like win rate analysis by outreach method, or understanding which activities most reliably move deals forward.

How Do Contacts, Companies, Deals and Activities Connect?

This is where the CRM database starts to feel less like a collection of separate forms and more like a living picture of your business relationships.

Here’s a concrete example. Imagine your rep, Sam, is working a deal with a software company called Meridian Labs. Here’s how that would look in a CRM database:

• There’s a company record for Meridian Labs with firmographic details – 200 employees, Series B funded, B2B SaaS
• There are two contact records linked to Meridian Labs: the VP of Sales who is the economic buyer, and the RevOps manager who is the technical evaluator
• There’s a deal record called “Meridian Labs – Enterprise Plan” valued at $48,000, currently in the “Proposal Sent” stage, with a close date of December 15
• There are twelve activity records attached to that deal – two discovery calls, one product demo meeting, eight email threads, and a note Sam wrote after the last call

When Sam’s manager opens the deal record, she can see the full timeline without asking Sam to brief her. When Sam goes on holiday and a colleague needs to cover, they’re not starting from zero. That’s the actual value of a well-maintained CRM database – institutional knowledge that’s accessible to whoever needs it, whenever they need it.

For teams thinking about how this structure feeds into RevOps reporting and forecasting, the relationship between these four record types is essentially the foundation everything else is built on. Get the structure right and the reporting follows. Build on messy data and even the best dashboards will mislead you.

Common Mistakes Beginners Make With CRM Data

Knowing what the records are is one thing. Using them consistently is another.

The most common early mistake is treating the CRM as a place to dump data rather than maintain it. Contacts get created with no company link, deals get opened with placeholder close dates that never get updated, and notes don’t get logged because logging them takes two minutes and nobody’s watching. Within a few months, the database reflects what your pipeline looked like at setup, not what it looks like now.

A second mistake is duplicating records. Two reps both create a contact for the same person at a company, and now you have split activity histories and two deals that should be one. Most CRMs have duplicate detection tools, but they only help if someone’s paying attention.

A third issue – especially for smaller teams – is under-using the company record. If you’re only filling in contact records and skipping the company layer, you’re making it much harder to understand your relationship with an account as a whole. That gap becomes painfully obvious the first time you try to calculate something like Customer Lifetime Value (LTV) by account.

If you’re evaluating which CRM to start with, the CRM Tools Directory is a good place to compare how different platforms handle these core record types – because the structure and flexibility vary more than you’d expect.

What to Prioritise When You’re Just Getting Started

If this is all new, don’t try to build a perfect CRM database on day one. It won’t happen, and the pressure to get everything right immediately is what causes teams to procrastinate on setup entirely.

Start with the basics. Get your company records in, link contacts to them, and create deal records for every active opportunity with a realistic close date and deal value. Log activities as they happen, even if it’s just a one-line note. That minimal structure is infinitely more useful than an empty system or a spreadsheet.

As you get comfortable with the fundamentals, you can start looking at how your CRM handles things like Monthly Recurring Revenue (MRR) tracking, custom fields, and pipeline reporting. But the foundation – contacts, companies, deals, activities – has to come first. Everything else depends on it.

For more guides on building out your CRM setup step by step, the CRM Guides section is worth bookmarking. And if you want to stay current on how CRM tools are evolving – including how AI is starting to auto-log activities and enrich contact records automatically – the CRM Daily Newsletter covers those developments as they happen.

The honest open question here is one that every growing team eventually faces: at what point does maintaining accurate CRM data become its own job? As databases scale into thousands of records, the cost of data hygiene – who owns it, how it’s enforced, what gets automated – is a real operational decision. There’s no universal answer, and the right one depends heavily on how your team is structured and how much your pipeline reporting actually drives decisions.