The enquiry came in on a Thursday. You meant to reply properly on Friday, but a client call ran long. By Tuesday, the person who asked had probably found someone else. You never knew, because their details were in row 47 of a spreadsheet that also contained seventeen other names you had not touched in two months. That is not a follow-up problem. That is a database problem, and fixing it does not require expensive software or a data analyst. It requires knowing which fields actually matter, which rules keep things clean, and what a working structure looks like before you have 200 contacts, not after.
A CRM database is the foundation underneath every lead management system, every follow-up reminder, and every sales pipeline. Get the foundation wrong and the tools on top of it fail regardless of how good they are.
What is a CRM database, and what does it actually store?
A CRM database is a structured record of every person and organisation that has contacted your business, along with the history of what was said, what was offered, and what happened next. It gives you a single place to track where each lead stands so nothing gets forgotten.
The word "database" sounds technical. In practice, a CRM database is closer to a very well-organised address book combined with a case file. Each entry - called a record - has fields: named slots where a specific type of information lives. A field for the person's name. A field for their email address. A field that records which pipeline stage they are in. A field that shows when someone last spoke to them.
What makes a CRM database different from a spreadsheet is not the fields themselves. A spreadsheet can have fields too. The difference is behaviour. A CRM database enforces relationships between records. It logs a history of every action taken. It can trigger a follow-up reminder when a certain number of days pass with no contact. A spreadsheet just sits there. The data in it does exactly what you do to it, nothing more.
The fields are only as useful as the rules that govern them. A CRM database with 30 fields where 20 are blank is worse than a database with 7 fields that are always complete. Incompleteness is the most common database failure in small teams, and it happens not because people are careless but because no one decided upfront which fields were mandatory.
For lead tracking purposes, the record needs to answer five questions at a glance: who is this person, how do I reach them, where did they come from, where are they in my pipeline, and when do I need to contact them next. Everything else is optional until you have a specific operational reason to collect it.
Which fields should every CRM database include from the start?
Every CRM database needs, at minimum, a full name, a primary contact method (email or phone), the source of the lead, the current pipeline stage, the date of the last contact, and a next follow-up date. Everything else is optional until you have a clear reason to collect it.
Here is a starter field set that works for a service business or agency with one to five people handling sales:
Contact fields:
- Full name (required, never split into first/last at the start - you can split later)
- Email address (required, used for deduplication)
- Phone number (optional at intake, but important for high-value leads)
- Company name (required if you sell B2B)
- Job title or role (optional, add it when it affects how you talk to them)
Lead and pipeline fields:
- Lead source (required - where did they come from? Website form, referral, social, event?)
- Pipeline stage (required - see the section below on how many stages to use)
- Lead value estimate (optional, but worth adding once deals are above a certain size)
- Lead qualification status (required if you are screening enquiries before engaging fully)
Activity fields:
- Date enquiry received (required, for measuring lead response time)
- Date of last contact (required, updated every time you speak or write to them)
- Next follow-up date (required, the single most important field for preventing lost deals)
- Notes (required, free text, a plain summary of the last conversation)
That is 12 fields. Most small-team CRMs have 40 or more by default. The discipline is in switching off the ones you do not use, not in filling them all in. An empty field is not neutral. It is noise. It makes you hesitate every time you open a record because you are not sure whether the field is blank because it is not applicable or because someone forgot.
Start with these 12. Add a field only when you have a specific decision that depends on it.
How many pipeline stages should a small business use?
Three to five pipeline stages is the right range for a small team. Enquiry, Proposal Sent, and Won or Lost covers most service businesses. Adding more stages only makes sense when each one has a clear action attached to it, not just a label.
A pipeline stage is a position in the sales process. It answers the question: what is the current status of this deal? The temptation, especially when setting up a new CRM for small business use, is to build a detailed pipeline that mirrors every step of your process. Fourteen stages. Colour coded. Very satisfying to design, very hard to maintain.
The problem with too many stages is not aesthetic. It is practical. If a stage does not have a clear entry condition and a clear next action, your team will disagree about which stage a lead belongs in. Records drift. The pipeline becomes fiction.
Here is a starter pipeline structure for a service business:
- New enquiry - someone has made contact, no conversation yet
- Contacted - you have spoken or written to them at least once
- Proposal sent - you have submitted a quote or scope
- Negotiating - they are interested, terms are being discussed
- Won - the deal is confirmed
- Lost - they chose not to proceed (keep these records - you may want to follow up later)
- Not qualified - the enquiry was not a fit, no deal possible
Keep Won, Lost, and Not Qualified as separate closed stages. If you merge them or delete them, you lose the data that tells you where your pipeline leaks.
The email follow-up timing you set at each stage will determine whether a contact advances or goes cold. Most small businesses set a follow-up reminder immediately after moving a lead to Proposal Sent, then another at three days, then another at seven. Research from XANT (formerly InsideSales) on lead response patterns - which measured response rates across millions of calls and emails rather than purchase behaviour - consistently shows that persistence across multiple contacts matters far more than any single message.
What are the most common CRM database hygiene mistakes, and how do you fix them?
The most common CRM database hygiene mistakes are duplicate records, missing lead source data, and stale contact details that nobody has updated. Each one compounds the others, and together they make contact management unreliable within months of setup.
Duplicate records happen when a lead comes in through two channels and gets entered twice, or when someone's name is spelled differently on two occasions. Duplicates are dangerous because they split the contact history. You look at one record and see that nobody has been in touch. Meanwhile, a colleague has been emailing the other version of the same person for two weeks.
Prevention is simple: designate the email address as your primary unique identifier and check against it every time a new record is created. If your CRM does not enforce this automatically, run a manual sort on the email column once a month. Duplicates compound quickly.
Missing lead source data is the second most common failure. People know where a lead came from at the moment of intake. Three weeks later, nobody remembers. If you do not capture lead source at the point of lead capture, you will never know which channels produce the deals that close, and which produce enquiries that go nowhere. This matters for any business deciding where to spend time or money on lead generation.
Stale contact data is the slow, invisible problem. People change jobs, phone numbers, and email addresses. Various sales data providers estimate that contact records decay at roughly 25 to 30 percent per year, though this varies heavily by sector and the figure is drawn from vendor research rather than independent studies. The practical takeaway is that a contact record untouched for 12 months is probably wrong about something. A quarterly pass through your inactive records - anyone with no activity in 90 days - and a decision to re-engage or archive them will keep your database accurate.
Bold the key fix: a monthly 20-minute tidy is easier than a quarterly three-hour rescue. Set a recurring calendar event. Open the CRM. Sort by last-modified date, oldest first. Fix the obvious blanks. Merge the obvious duplicates. Archive the clearly dead leads. Twenty minutes, once a month, is enough to maintain a database of up to 500 contacts cleanly.
How do you set up a CRM database that stays clean at 1,000 contacts and beyond?
The transition from manageable to chaotic usually happens somewhere between 300 and 600 contacts, when the database is too large to scan visually but too small for anyone to feel the problem is urgent. The answer is rules written down before you reach that number, not after.
Mandatory fields at entry are the first rule. Decide which fields are required before a record can be saved. At minimum: name, email, lead source, and pipeline stage. If your CRM does not enforce this technically, make it a team agreement and check compliance weekly until it is habit.
A defined lead qualification process is the second rule. Not every enquiry deserves a full record with active follow-up. Some people are browsing. Some are competitors. Some are students researching for an essay. A quick qualification check - does this person have a real need, a real budget, and a real timeline? - determines whether a new enquiry gets a full record and active sales follow-up or gets filed as Not Qualified and archived.
Kodeleads handles both of these by prompting for source and stage at the point of capture, so the data is there from the first moment rather than being retrofitted.
A regular deduplication check is the third rule. Monthly for smaller databases, weekly if you are generating more than 20 new leads per week. Most CRM tools have a built-in duplicate finder. Use it on a schedule rather than waiting until the problem is visible.
An archiving policy is the fourth rule. A closed deal - won or lost - should not sit in your active pipeline view forever. Set a rule: any contact with no activity in six months goes to an archived view, not deleted, but out of the daily working list. Deletion is rarely the right choice. Archiving keeps the database clean without destroying the history.
The deeper point is this: a CRM database is not a filing system. It is a decision-support system. Every record should answer the question "what should I do next with this person?" If a record cannot answer that question - because the fields are blank, the stage is wrong, or the last-contact date is two years old - it is not a record. It is clutter.
A clean database at 1,000 contacts makes every sales follow-up faster, every email follow-up more relevant, and every lead nurturing decision more confident. A dirty database at 200 contacts makes all of those things slower, because you cannot trust what you are looking at.
Frequently Asked Questions
What is a CRM database?
A CRM database is a structured record of every person and organisation that has contacted your business, along with the history of what was said, what was offered, and what happened next. It gives you a single place to track where each lead stands so nothing gets forgotten.
What fields should every CRM database include?
Every CRM database needs, at minimum, a full name, a primary contact method (email or phone), the source of the lead, the current pipeline stage, the date of the last contact, and a next follow-up date. Everything else is optional until you have a clear reason to collect it.
How often should you clean a CRM database?
A light clean every month (fixing obvious duplicates and blanks) and a deeper review every quarter is enough for most small teams. At 1,000 or more contacts, a quarterly deduplification pass and a six-month decay check on inactive records will keep the database usable.
What is data decay in a CRM database?
Data decay is the natural process by which contact records become inaccurate over time. People change jobs, phone numbers, and email addresses. Research from various sales data providers suggests roughly 25 to 30 percent of contact records become outdated within a year, though the exact figure depends on your industry.
How many pipeline stages should a small business use in their CRM?
Three to five pipeline stages is the right range for a small team. Enquiry, Proposal Sent, and Won or Lost covers most service businesses. Adding more stages only makes sense when each one has a clear action attached to it, not just a label.
What is the difference between a CRM contact and a CRM lead?
A lead is someone who has expressed interest but has not yet been qualified. A contact is a broader term for any person stored in the database, which can include leads, existing customers, past customers, and suppliers. Most small-team CRMs use the terms interchangeably, but it helps to know where the distinction matters for your own pipeline stages.
How do you prevent duplicate records in a CRM database?
The most reliable prevention is a single mandatory field used at point of entry - usually email address - that the system checks against existing records before saving. If your CRM does not do this automatically, run a manual sort on the email column once a month. Duplicates compound quickly and are far easier to prevent than to merge later.
Does a small business need a CRM database or will a spreadsheet do?
A spreadsheet works for roughly the first 50 to 100 contacts if one person owns it and updates it every day. Beyond that, the absence of follow-up reminders, the lack of a contact history log, and the ease of accidental overwrites make a purpose-built CRM database the more reliable choice, even a simple one.
Try Kodeleads if your database is already getting away from you
If your contacts are split across an inbox, a spreadsheet, and someone's memory, Kodeleads gives you a clean starting structure - the right fields, enforced pipeline stages, and automatic follow-up reminders - without the setup overhead of a tool built for a 20-person sales team. Try Kodeleads