A lead arrives in your inbox. That is where most website advice stops, and it is where the work actually starts.
The word CRM makes people picture software with a login screen, so the question becomes which one to buy. That is the wrong first question. A CRM is a shape, and you can hold that shape in a spreadsheet or in a system that costs more per month than your hosting. Getting the shape right first is what stops you paying for a product nobody in your company opens.
What a CRM actually is#
Four records. A person. Everything that has happened with them. Where they sit in your process. What happens next, and when.
That is the whole thing. Every feature on every CRM comparison page is an addition to one of those four, and most additions are worth nothing until the four underneath them are being kept honestly.
So the test is not which software you have. A spreadsheet with those four columns, kept current, is a CRM. Software with two hundred fields where nobody records what happens next is a filing cabinet with a subscription.
The handoff is the part that breaks#
The path is short: someone fills in a form, a notification goes somewhere, a record gets created. Almost nobody loses leads at the form. They lose them between the notification and the record, because the notification is an email and email is where things go to be meant to be dealt with later.
You can recognise this without any tooling. Two people give different answers about whether a customer was called back. A quote gets sent twice. Somebody asks what happened to the enquiry from the trade show and the honest answer involves searching an inbox.
Pick an enquiry from three weeks ago. Without searching your email, say who it was, what they wanted, what you sent them, and what is supposed to happen next. If you cannot, you have lead capture and not a CRM, and no amount of software fixes that until the handoff does.
Getting the lead as far as the inbox is a separate job, and the lead-generating website guide covers it. This article starts where that one ends.
Automated capture, and what automated has to mean#
Automated means the record exists without anybody deciding to create it. If a person has to copy an email into a system, that step will be skipped on the days you are busiest, which are the days the leads arrive.
Capture three things at the moment of the form, because they are free then and expensive later: which page they were on, what they actually typed, and when. The page tells you which service they were reading about, which is most of what a first reply needs to sound informed.
Resist the urge to ask for more. Every required field is one more reason to close the tab, and the fields people add at this stage are usually the ones a two-minute phone call answers better. You can test this on your own form: add a field, watch a month, take it away again.
Customer profiles#
A profile is a person, not a company. Companies do not answer the phone. Keep it small enough that it stays true:
- Name and how to reach them, with one channel marked as the one that works.
- Where they came from, in the plain sense of which page or which referral.
- What they asked for, in their words rather than your categories.
- What you have sent them, and when.
- What they bought, if anything, and what it was worth.
A field that nobody fills in is worse than a missing field, because it makes the record look complete. If half your profiles have an empty industry dropdown, the answer is to delete the dropdown.
The pipeline is a promise about what you do next#
Stages should describe your behaviour, not the customer's mood. "Interested" and "Warm" are unfalsifiable and quietly become storage. A stage earns its place when there is one obvious thing that moves a record out of it.
| Stage | What it means | What moves it on |
|---|---|---|
| New | They asked, nobody has replied | Your first reply goes out |
| Contacted | You replied, the ball is with them | They answer, or the follow-up date arrives |
| Quoted | A number is with them | They accept, they decline, or the chase date arrives |
| Won | They bought | Nothing. It is a record now |
| Lost | Closed without a sale, with a reason | Nothing, and that is the point |
Four or five stages is enough for most businesses. If a stage sits empty for a month, delete it. If a stage holds everything, it is doing the work of two and needs splitting.
Tasks and follow-ups are the part that changes the money#
Of the four records, this is the one that pays for the other three. Profiles and history are a description of the past. A next action with a date is the only part that causes something to happen.
One rule covers it. Every open lead has a next action and a date, or it is closed. Closing a lead is a decision and a fine one. Drifting is not a decision, and drifting is what most pipelines are full of.
A lead you have decided not to chase, marked lost with one line about why, costs you nothing and teaches you something at the end of the quarter. The same lead sitting in Contacted for a year costs you attention every time you look at the list, and teaches you that the list is not worth looking at.
Quotes and deals#
Whether a quote should be generated by software or written by a person is a separate decision, and it is covered in the guide on adding business features. What matters here is the record it leaves behind.
A deal is a number, a date, and a state. Keep one deal per thing you are selling rather than one per customer, because a customer who buys twice a year is a good customer and a single row that gets overwritten twice a year is a lost history. It also makes the obvious question answerable: what did we quote, to whom, and how much of it came back.
History is what makes the second sale easy#
The first sale is won by the website and the reply. The second is won by remembering. Someone who has to re-explain their situation to the company they already bought from is a customer you are about to lose to a competitor who takes notes.
The practical minimum is one line per interaction: the date, the channel, and what was said or agreed. Not a transcript. The point is to make the next conversation start in the right place, and one honest sentence does that better than a paragraph nobody will read.
One system, not two#
Two pieces of software is fine. Two sources of truth is not. The website is where a lead is created and the CRM is where it lives, and the only thing that has to be true is that nobody has to check both to answer a question.
There are three levels of connection and they suit different sizes of business. The form emails you and somebody types it in, which works and does not scale. The form plugin writes to the CRM directly, which is configuration and covers most cases. A built integration, when the CRM is the system you already run the business on and it has to agree with your website about who a customer is.
Worth saying plainly, because the alternative is a reader who goes looking: Sitebefy builds the website, not the CRM. There is no CRM in your dashboard. Your site is an ordinary WordPress installation, so a form plugin can push into whichever CRM you choose, and when no ready-made connection fits, the join between the two is custom development: described, quoted, and billed separately from your subscription, then shipped onto the site you already have.
And what gets commissioned is almost never a CRM written from scratch. Contacts, stages and follow-ups have been built thousands of times and you should buy them. What has not been built is the join: the rules that decide when a website enquiry becomes a customer record, which system wins when the two disagree, and how your own workflow maps onto somebody else's idea of a pipeline. That part comes from how your company operates, which is exactly the test for paying a developer at all.
Build only what the business needs#
CRM projects fail on adoption far more often than on features. A tool nobody updates is not a partial CRM, it is a wrong answer to every question you ask it, which is worse than having no answer at all.
- Start with the four records and nothing else. A shared sheet is a legitimate first version.
- Automate the capture, so records appear without anybody remembering to make them.
- Add stages only when you can name what moves a record out of each one.
- Add a field only after the third time you wished it existed.
- Move to real software when the sheet has stopped being trustworthy, and not before.
Who updates this, and on what day? A field with no answer to that will be empty in a month, and an empty field on a record that looks complete is how people end up trusting a system that is quietly wrong.
This is the same discipline that applies to the website itself. Reporting dashboards, lead scoring, sequences and territory rules are all real things that real companies need, and none of them are what your first version needs. They are what you buy after a year of records that people actually kept.
Start with a website your CRM can talk to
Describe what you do and Sitebefy builds a real WordPress site you own, with the pages, layouts, content and SEO already in place. Because it is ordinary WordPress, your forms can push straight into whichever CRM you pick, and a developer can build the integration later if the ready-made one does not fit. Look at the result before creating an account, and every plan carries a 14-day money-back guarantee.
Build my siteA week's worth of CRM#
- Make one sheet with five columns: person, what they asked, stage, next action, date.
- Put every open enquiry from the last month into it, including the ones you feel bad about.
- Give each one a next action and a date, or mark it lost with a reason. Both are progress.
- Connect the form so the next one lands there without you.
That is a working CRM, and it took an afternoon. What you learn running it for a month is the specification for whatever you buy or build next, and it will be a shorter specification than the one you would have written today.
The companies that get value out of expensive CRM software are almost always the ones that were already keeping the four records by hand. The software made them faster. It has never made them exist.
