How do I choose a CRM for a service business?
Start by deciding what the system must be the source of truth for. If the answer is jobs, dispatch and invoices, you need field service software, not a CRM. If it is marketing follow-up, nurture and pipeline, you need a sales CRM. Most operators end up running both, and the decision that actually matters is which one owns the customer record and how the two stay in sync.
Three different products all called CRM
The word covers at least three categories that solve different problems, and demos rarely make the distinction clear.
Buying the wrong category is expensive not because the software is bad but because you end up bending it into a shape it was not designed for, then paying someone to maintain the bend.
- Field service platforms. ServiceTitan, Jobber, Housecall Pro and similar. Built around jobs, capacity, dispatch and invoicing. The customer record exists to support the work order.
- Marketing and sales CRMs. GoHighLevel, HubSpot and similar. Built around pipelines, sequences, campaigns and conversations. Strong on outbound, weak on scheduling real crews against real capacity.
- General-purpose CRMs. Endlessly configurable, which means someone has to configure them and keep configuring them. Powerful if you have that person, a liability if you do not.
Write down what each system must own
Before you look at features, list the facts your business depends on and assign each one exactly one owner: customer identity, service history, revenue, marketing consent, lead source, conversations, capacity.
Two systems is normal. Two systems that both believe they own revenue is a permanent reconciliation tax. The point of an integration approach is to let each platform be authoritative for what it is good at, and to keep the copies downstream honest and clearly labeled as copies.
Questions that predict pain eighteen months out
Feature checklists age badly. Structural constraints do not.
- Can I get my data out, in full, on my schedule? Not a summary report — the underlying records, including custom fields and historical edits.
- Is there a documented API, and what are the rate limits? Ask for real numbers. Low limits shape everything an integrator can build for you later.
- Are there webhooks, or only polling? Webhooks make near-real-time workflows possible. Polling makes them expensive.
- How are custom fields handled? Whether they appear in the API, whether they are typed, and whether they survive an upgrade.
- What happens to history when a record is edited or deleted? Systems that overwrite silently will corrupt any trend you try to build.
Fit to your actual operating model
A two-truck shop and a fifteen-location operator have opposite failure modes. The small shop dies from complexity and abandoned data entry. The multi-location operator dies from inconsistency — every branch using fields differently until rollup reporting is impossible.
If you are running several locations, decide early whether you want one tenant with business units or separate instances, because that choice is painful to reverse. And be honest about who will administer the system. Software adoption is a staffing question disguised as a purchasing question — see how the customer record gets used day to day before you commit.
Topics: choosing a CRM · evaluation · system of record · buying
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.