CRM implementation
Most CRMs aren't broken. They're configured for a business that isn't yours.
Forty required fields, stages nobody recognises, and a report nobody opens. A CRM works when using it is easier than not using it — and that is a design problem, not a licensing one.
HubSpot and Salesforce · Configured from watching how you sell · Licences stay in your name
The first fortnight
Days 1–2
Watch a sale happen
A ride-along, not a workshop. The pipeline comes from what we see.
Days 3–5
Build the pipeline
Five stages people recognise, and the fields that support reporting.
Days 6–8
Migrate and connect
Existing data in, phones and forms wired to it.
Days 9–14
Roll out in stages
Training that survives the first busy week, which is the real test.
What does a CRM implementation actually involve?
Designing a pipeline that matches how the business really sells, capturing leads automatically with their source, keeping the required fields to the minimum that supports reporting, migrating existing data, connecting the website, phones and ad platforms, and rolling it out in stages with training that outlasts the first busy week.
The technology is the easy part. Every mainstream CRM can do what a service business needs. What decides whether it works is whether the person quoting a bathroom on a Tuesday finds it faster to use than to not use — and that is decided by how many fields they have to fill in and whether the system gives them anything back.
Which is why implementations here start by watching a sale happen rather than by opening the software. A pipeline designed from an org chart produces stages nobody recognises; a pipeline designed from a ride-along produces stages people can point at.
A pipeline that matches how you sell
- EnquiryEvery source lands in one placeAutomatic
- ContactedThe stage most CRMs never record honestlyAutomatic
- Appointment bookedThe number a service business is actually run onManual
- QuotedWith the value attached, so forecasting means somethingManual
- Won or lostLost reasons captured, because that is the useful halfManual
The shape
Five stages, each one an event you can point at
New enquiry
Created automatically from the form, the call or the marketplace. Source captured at creation, because nobody ever fills it in later.
Contacted
Someone has actually spoken to them. The gap between this stage and the one above it is where most lost revenue is measurable.
Site visit booked
For most trades this is the real conversion event. Cost per booked site visit is a far more useful number than cost per lead.
Quoted
Value and scope on the record. Without this the business cannot calculate a close rate, and without a close rate no ad budget can be justified.
Won / Lost
With a reason on the loss, chosen from a short list. Six months of loss reasons is the most useful sales document a service business can own.
Five stages, not eleven. Every additional stage is another place a record can go stale and another judgement call for whoever is updating it. If two stages are always updated at the same moment, they are one stage.
What's included
What implementation covers
A pipeline that matches how you sell
Stages named after things that actually happen — site visit booked, quote issued — not a generic template. If a stage does not correspond to an event someone can point at, it gets removed.
Lead capture with the source attached
Forms, calls, chats and marketplace enquiries creating records automatically with the source, campaign and service captured at creation, because nobody backfills it later.
Only the fields that get used
Most failed CRM rollouts died of forty required fields. We start with the smallest set that supports the reporting you need, and add only when something is demonstrably missing.
Routing and ownership rules
Every record has an owner from the moment it exists, assigned by service, territory or rota, with escalation if it sits untouched past your agreed window.
Quote and job values on the record
So close rate, average job value and cost per booked job are calculable rather than estimated. This is the difference between a CRM and an expensive address book.
Reporting a sales manager will open
One dashboard: what came in, what got contacted and how fast, what got quoted, what closed, and why the rest did not. Not eighteen widgets nobody has looked at since launch.
Migration from the spreadsheets
Existing customer and lead data cleaned, deduplicated and imported — including the shadow spreadsheet on somebody's laptop, which is usually the most complete record you have.
Training that survives the first busy week
Short recorded walkthroughs of the four things people actually do daily, plus a written process. A two-hour launch session is forgotten by the following Tuesday.
The number nobody has
Six months of loss reasons is worth more than any dashboard
Ask most service businesses why they lose quotes and you get an answer about price. Ask the CRM, after six months of a required loss reason chosen from a short list, and the answer is usually different: we took too long to get back to them, they went with someone who could start sooner, the scope was not what they wanted.
Those are fixable. "We were too expensive" is not, and it is what everyone assumes in the absence of data. One dropdown, filled in at the moment a quote is marked lost, is the cheapest piece of business intelligence a service company can install — and it is one of the first things we configure.
How it works
Watch, design, build, roll out narrow
Watch a sale happen
Week 1Sit with the people who sell and see how a job actually gets won, including the bits that are not in anyone's process document. A CRM configured from an org chart rather than from observation never gets used.
Design the pipeline and the minimum fields
Week 2Stages that map to real events, the smallest useful field set, and agreement on what "contacted" means. That last one causes more reporting arguments than anything else.
Build, migrate and connect
Weeks 3–6Configuration, data migration from spreadsheets and the old system, and connections to the website, phones and ad platforms so records create themselves.
Launch narrow, then widen
Weeks 6–10One team or one service line first, for a few weeks, with the process corrected from what actually goes wrong. Then everyone. Big-bang CRM launches are how businesses end up back in the spreadsheet.
Fit
Who this suits, and who it doesn't
This is a good fit if
- You cannot currently say what your close rate is without estimating.
- Leads live in an inbox, a phone and a spreadsheet, and nobody owns them formally.
- The CRM was configured by whoever set it up years ago and nobody has revisited it since.
- You want to push closed-won values back into your ad platforms.
- Someone internally will own the system after we hand it over.
This isn't the right fit if
- Nobody has agreed who owns a lead. Software does not settle that argument, it exposes it.
- You want every field made mandatory so the data is complete. That is how you get a CRM full of "n/a".
- You are buying a CRM because a competitor has one. That is not a requirement.
- The sales team has been told a CRM is coming to monitor them. Adoption will fail and no configuration fixes it.
Related
What a CRM connects to
Lead response & follow-up
Enquiries answered in minutes at any hour, quotes chased for a month, reviews requested automatically. Email sequences included.
Workflow automation
Your CRM, phones, ads, quoting and accounting wired together so nobody retypes data between them.
Lead generation
Every part above, run together as one path from first click to booked job, and reported as one cost.
Further reading: how a CRM helps a small business follow up with more leads.
Questions about CRM implementation
Which CRM should a service business use?
HubSpot suits most small and mid-sized service businesses: usable, good automation, sensible pricing at the low end. Salesforce is right where you have complex processes, multiple divisions or existing investment. Both are fine — configuration decides whether a CRM works far more than the brand does.
Why do CRM rollouts fail?
Almost always adoption, not software. Too many required fields, stages that do not match how people actually sell, no owner, and training that happened once. If entering data costs a salesperson more than it gives them back, they will keep using their phone and a notebook.
Do we need to replace our current CRM?
Usually not. Most CRMs in service businesses are capable and badly configured. Reconfiguring is faster, cheaper and much less disruptive than migrating. We will only recommend a move where the platform genuinely cannot do what the business needs, and we will show you why.
How long does an implementation take?
Six to ten weeks for a typical service business, including migration and a staged rollout. It can be done in two if you skip observation and training, and those are the implementations that get abandoned in month four.
What about the data in our spreadsheets?
It gets cleaned, deduplicated and migrated. The shadow spreadsheet is usually the most complete customer record in the business, precisely because it was maintained by whoever needed it to work. It is an asset, not an embarrassment.
Will the team actually use it?
Only if it gives them something. The fastest route to adoption is making the CRM the place their leads arrive and their follow-ups are prompted, so ignoring it costs them work. Requiring data entry that only benefits management is how you get a system full of blanks.
Can you connect the CRM to our ads and phones?
Yes, and that is where most of the value is. Calls logged against records, forms creating leads with the campaign attached, and closed-won values pushed back into Google and Meta so bidding optimises against revenue rather than form fills.
What does CRM implementation cost?
Quoted as a project, separately from the monthly range, depending on data migration volume and how many systems connect to it. Licence costs are yours and paid directly to the vendor. Ongoing optimisation, if you want it, sits inside a monthly retainer.
Start with one question: what's your close rate?
If the answer involves an estimate, the CRM is not doing its job. A thirty-minute review will tell you whether that is a configuration problem or a process one — they need different fixes.
You talk to Mathew.Not an account manager, and not a sales team.