How a CRM works
A CRM — customer relationship management system — is the single place where every lead and customer exists as a record. A contact arrives from a form, a phone call, WhatsApp or a spreadsheet import; it is stored once, along with the source that produced it; and from then on every quote, call, email and change of status is written against that record rather than living in one person’s inbox or notebook.
Two ideas do most of the work. A pipeline is the ordered list of stages a deal passes through — new enquiry, contacted, quoted, won or lost — so at any moment you can see how much work is sitting at each stage and what is stuck. Ownership means one named person is responsible for each record, which is what stops a lead being chased twice by different people, or not chased at all because everyone assumed somebody else had it.
Why a CRM matters
For marketing, the CRM is where a lead stops being a form submission and becomes a number you can trust. Cost per lead is easy to improve by buying cheaper leads; cost per customer is not, and you cannot calculate it unless outcomes are recorded somewhere consistent. The CRM is that somewhere, and without it every claim about return on ad spend rests on somebody’s memory.
It also makes the ad platforms cleverer. When you can tell Google or Meta which enquiries became customers, and which turned out to be junk, bidding starts optimising towards the enquiries that pay rather than towards whichever form is easiest to complete. That feedback loop is impossible without a system of record, which is why it is usually the first thing I ask about before increasing anyone’s budget.
Common mistakes with a CRM
Buying a platform before agreeing a process is the usual failure. Stages nobody updates, fields nobody fills in, and a sales team still working from WhatsApp produce a CRM that is expensive and wrong at the same time. Agree the stages, define what each one means, and name who moves a record between them. Then choose the software, which is the easy part.
Two smaller habits cause most of the reporting pain. Losing the source: if an enquiry lands without the campaign that produced it, every marketing report afterwards is guesswork dressed up as data. And never marking losses: a pipeline in which nothing is ever closed as lost tells you nothing, because the wins have no denominator to be measured against.
What to do about it
Keep the first version small. Contact details, source, stage, owner, value, and a next action with a date against it. That is enough to run a pipeline and to report honestly, and it is few enough fields that people will actually complete them. Add more only when a specific question you keep having to answer requires it.
Then close the loop in both directions. Pass campaign details through from the website form into the record so the source survives the handover, and send outcomes back out to your ad platforms and reports as they are decided. That wiring is a technical job rather than a sales one — it is what CRM integration means in practice — and it is worth finishing before you spend more on traffic. Where deals close weeks after the click, the mechanism for reporting them is an offline conversion import.
