CRM best practices: a worksheet for the CRM you already have
Most CRM advice tells you how to pick one. That is rarely the problem. The problem is the CRM you already bought, half filled with stale records nobody trusts. This worksheet audits the system you have, in the order that actually recovers lost leads. Seven checks, and the first one usually explains the rest.
You probably don't need a different CRM
When someone tells me their CRM is not working, they usually mean one of two things. Either the data in it is wrong, or nobody on the team believes the data in it. Those are the same problem.
Once sales stops trusting the system, they start working around it. A spreadsheet here, notes in a phone there, deals tracked in someone's head. Now the CRM is not just wrong, it is confidently wrong, and every report built on it is fiction.
Buying a new one does not fix that. It moves it. So before anyone evaluates software, work this list against what you own.
Seven checks, in order
Is there one source of truth?
Count the places customer information currently lives. Be honest.
- The CRM
- The shared inbox
- The quoting spreadsheet
- Someone's notebook or phone
If the answer is more than one, that is your first fix and nothing below matters much until it is done. Pick the system of record, move everything into it, and close the alternatives. Not deprecate. Close.
What percentage of records are junk?
Pull the full contact list and count:
- Exact duplicates, and near-duplicates with a typo in the email
- Records with no email and no phone
- Contacts at companies that no longer exist
- Anything untouched in two years with no reason to keep it
Most owner-led businesses find a fifth to half the database is dead weight. A smaller list you trust beats a big one you do not.
Which fields are actually filled in?
Go field by field. For each one ask two questions: what decision does this field inform, and what percentage of records have it populated?
Any field under half filled that informs no decision should be deleted. Every optional field is a small tax on every person entering data, and the tax is paid whether or not the field is ever used. Keep what drives routing, follow-up, or reporting. Cut the rest.
Do you know where leads come from?
- Is there a lead source field, and is it required
- Is it a fixed list, or can people type whatever they want
- Do web form submissions populate it automatically
- Can you answer which source produced closed revenue last quarter without guessing
Free-text lead source is one of the most common findings I have. You end up with Google, google, GOOGLE, Web, and Website as five separate sources, and no usable report.
Does follow-up actually fire?
- Is there a defined first-response time, and does anyone measure it
- What happens automatically when a form is submitted
- What happens to a lead that goes quiet after one exchange
- Who owns a lead nobody has touched in 30 days
This is where most revenue leaks. Not at the top of the funnel. In the gap after someone raises their hand and before anyone follows up.
Does everyone agree what the stages mean?
Ask three people on your team to define your pipeline stages independently. If you get three different answers, your pipeline report is decorative.
Write one-sentence definitions for each stage, including the specific event that moves a deal forward. Then check that the stages in the system match the ones on the list.
Is there a report anyone actually reads?
- One report, weekly, that changes a decision
- Numbers the owner understands without translation
- Something that shows lead quality, not just lead count
If nobody opens it, it is not a report. It is a chore.
The lead volume was never the constraint
At a wire manufacturer we rebuilt the CRM completely and brought content and CRM work in-house rather than keeping it with an outside agency. Neither of the results below was a software decision.
Reduction in unqualified leads, alongside a paid search and messaging rebuild. The wrong half stopped arriving.
Qualified quote requests in year one, from a pipeline the team could finally forecast from because they trusted it.
At an electrical contractor, the first real CRM plus a repeatable follow-up process is what turned a busy phone into something you could plan around.
The whole worksheet is above. I'll send the version with scoring built in.
Before you start
What are CRM best practices for a small business?
One source of truth, clean and current data, a required and fixed-list lead source field, follow-up that fires automatically, pipeline stages everyone defines the same way, and one weekly report someone actually reads. Feature depth matters far less than whether the team trusts what is in the system.
How do I know if my CRM data is bad?
Pull the full list and count duplicates, records with no contact method, and anything untouched in two years. If more than a fifth falls into those buckets, your reporting is already unreliable. The other tell is behavioral: if anyone keeps a private spreadsheet, they have already told you they do not trust it.
Should I switch CRMs or fix the one I have?
Fix the one you have first, unless it genuinely cannot do something you need. Migrating dirty data into new software produces the same problems with a new login screen, plus a rollout to manage. Work this worksheet first, then decide.
How does CRM quality affect lead quality?
Directly. Bad data means routing errors, missed follow-ups, and duplicate outreach, all of which cost you leads you already paid for. It also hides which sources produce real revenue, so you keep spending on the ones that do not. Cleaner data is usually cheaper than more traffic.
Who should own the CRM?
One named person, not a committee and not whoever has time. They do not have to do all the data entry, but someone has to own the definitions, the required fields, and the standard. Shared ownership of a database reliably produces an unowned database.
A CRM is only worth what your team believes about it.
Most CRM problems are not really CRM problems. They are a symptom of the marketing problem that isn't a marketing problem: the website, the CRM, and the follow-up each work alone and drop the handoff between them. Cleaning the database helps. Connecting the systems is what stops the leak. If the worksheet turns up more than you want to deal with, the $499 marketing audit reads across all of it and hands you a prioritized order to fix it in.