CRM Diagnostics
A CRM health check that tells you what is actually wrong.
A fixed-scope CRM audit of what you run today, on Salesforce, HubSpot or Odoo. Thirty-six checks across six domains, a scored report, a prioritised fix list, and a readout with the people who will do the work.
Read-only access. Ten working days on a single platform. One diagnostic, whichever of the three you run.
Before the diagnosis
Four different problems that all look like a CRM problem
Almost every CRM complaint turns out to be one of four things. They need different fixes and they cost very different amounts. No demo will tell you which one you have.
The process was never written down
Two teams describe the same pipeline stage differently. No amount of configuration settles an argument about what "qualified" means, because the argument is not about software.
The configuration drifted
The org was right for how the business worked two years ago. Fields accumulate, automations stack on top of each other, and nobody left can say why half of them exist.
The data was never trusted
Duplicates, blanks, three spellings of one company. Every report becomes a negotiation, so the forecast gets assembled by hand and defended in a meeting.
People never adopted it
The licences are paid and the logins are low. The real pipeline lives in a spreadsheet that one person maintains and nobody else can read.
This is why a health check comes before a purchase. Process work, reconfiguration, a data programme and adoption work are four different projects. Buy the wrong one and the symptom survives the invoice.
Scope
What a CRM audit covers, in full
Thirty-six named checks across six domains, published before you buy rather than revealed in a proposal. Measured against your process, not against a best-practice list written for somebody else's business.
Process and handovers
- Lead capture and routing rules
- Stage definitions, compared across teams
- Deal to order handover
- Case intake and assignment
- Approval steps and what they block
- Points where records get retyped elsewhere
Data quality
- Duplicate rate on accounts and contacts
- Required fields actually populated
- Orphaned records with no owner
- Picklist values still in use
- Ownership and reassignment history
- Data entered outside the CRM
Configuration and build
- Custom objects against standard ones
- Field count and unused fields
- Automation inventory: flows, rules, scripts
- Validation rules and conflicts between them
- Permission sets and the sharing model
- What last changed, and by whom
Integrations and data flow
- Every system the CRM talks to
- Direction and frequency of each sync
- Failure behaviour and alerting
- Field-level mapping collisions
- Rate limits and volume ceilings
- Who is paged when one breaks
Adoption
- Active users against paid licences
- Records created per user per month
- Features nobody touches
- Mobile and email usage
- How new joiners are onboarded
- Where the real work happens instead
Licensing and commercials
- Edition against features in use
- Seats purchased against seats active
- Add-ons and their justification
- Renewal dates and notice terms
- Storage and API consumption
- Overlapping or duplicate subscriptions
Each check comes back as pass, partial or fail, with the evidence attached. A fail with no evidence is an opinion, and opinions are not what you are paying for.
The deliverable
One written verdict, not a sixty-slide deck
An audit is only useful if the person holding it can make a decision. So the report is written for the decision-maker, not for the implementer.
WHAT LANDS ON YOUR DESK
A verdict you can act on, hand to a vendor, or file and ignore.
Every finding is tied to something you can open and check: a screen, a field, a report, a count of duplicate records, a licence line on an invoice. Nothing in it asks you to trust our judgement instead of your own evidence.
- All six domains scored, with the evidence behind each
- Findings ranked by impact, each with an effort and cost band
- The fixes worth doing in under a week
- What you can stop paying for at renewal
- A build-or-fix recommendation, stated plainly
- A readout session with whoever will own the work
If your setup is fine, the report says that. We would rather lose an implementation than sell one that was never going to work. That is not generosity, it is the only way an independent verdict means anything.
How it runs
Ten working days, and how much of your time it takes
A diagnostic that runs for two months has stopped being a diagnostic. Most of this happens on your data rather than in workshops, which is why the client time is counted in hours, not weeks.
| Stage | Duration | Your involvement |
|---|---|---|
| Access and interviews | 2 days | 1 hour with each team lead, plus read-only access |
| Data and configuration analysis | 4 days | None. We work inside your org |
| Integration tracing | 2 days | 30 minutes with whoever owns each integration |
| Scoring and write-up | 2 days | None |
| Readout | Half day | 90 minutes, with everyone who owns a fix |
What is not included, stated up front
- Remediation and reconfiguration
- Sandbox rebuilds
- Managed-package internals
- Penetration testing
- Bulk data cleaning
- Licence purchase
- User training
There is no licence markup in our advice, so the cheapest correct answer is also the one we are paid to give. If a finding needs building, that is a separate project with its own scope — the report will name it, not quote it.
Honesty about scope
When a health check is the wrong purchase
Three situations where an audit would just add ten days and a document.
You already know what is broken
If the cause is not in dispute and the only open question is who builds the fix, an audit is a delay. Go straight to implementation with a defined scope.
CRM implementationYou have not chosen a platform yet
Auditing a system you are about to leave is money spent twice. The right first engagement is a platform decision, which is a different piece of work.
CRM consultingIt went live last month
A three-month-old system with low usage usually has a training and change problem, not a design problem. Give it a quarter, then audit if it is still off.
A health check is the right first step when the symptom is real but the cause is contested. That describes most CRM projects that quietly stalled, which is why this is the most common way our engagements start.
Proof
What the same method found
Three engagements where the first useful output was a correct diagnosis, not a new platform.
Customer service and sales management at parcel-delivery scale
Enquiry volumes outgrew what the process could absorb. Mapping how enquiries actually arrived showed the break was in intake and routing, not in the platform.
Read case studyClient onboarding transformation for a licensed EMI
Onboarding had to satisfy regulatory requirements and stay workable for the team. Process and data were designed before a single field was configured.
Read case studyDonation and volunteer management for a national Red Cross society
Only around 60% of recurring donations were being processed. The cause was bank integration, not the CRM, so the fix was integration work and the platform stayed.
Read case studyTell us what is not working, and when it started
We will tell you whether a health check is the right first step, or whether a cheaper conversation answers it.
FAQ
CRM audits and health checks, answered
What is a CRM audit, and how is it different from a health check?
How long does a CRM audit take?
What does a CRM audit cost?
Why use one firm across Salesforce, HubSpot and Odoo rather than a single-platform specialist?
What access do you need, and is it safe?
Will you tell us to replace our CRM?
Do we have to hire you to fix what you find?
Start with the truth
Tell us what is not working, and when it started
We will tell you whether a CRM health check is the right first step, or whether a cheaper conversation would answer the same question. If it is not an audit, we will say so on the call.