Diagnosis · six symptoms, six causes
Something is broken.
Nobody can say what.
Forecasts are wrong, or sales stopped trusting marketing, and the honest answer inside the company is that nobody knows whether it is the data, the architecture, or two teams who never agreed a definition. Tick what you are seeing. You get the cause it usually turns out to be, a way to confirm that yourself in ten minutes, and only then a price.
Nothing here is sent anywhere. The checks are things you run in your own org, and you can act on all of them without hiring anybody.
Why the check comes first
Why not just quote the repair?
Because three of the six symptoms look identical from the outside and are not the same problem underneath. Quoting a fix before measuring is how consultants end up rebuilding scoring when the actual fault was sync, and how clients end up paying twice.
What does a €500 check produce?
Four measurements against reports your admin exports, a verdict of A, B or C, and a costed order of work with the dollar impact attached to each finding. Three days, read-only, no access to your org and no permission set created.
What does skipping it cost?
The common version: a team concludes scoring is broken, pays to rebuild it, and discovers six weeks later that prospects were never reaching Salesforce in the first place. The rebuild was competent and irrelevant. Measuring first is cheaper than being right by accident.
When is the answer to do nothing?
Roughly a third of these end with a recommendation that costs nothing, most often because the tooling is fine and two teams have never agreed what a qualified lead is. No consultant can fix that for you, and saying so is more useful than selling around it.
Order of work
Which problem should we fix first?
When several symptoms are present at once, the order matters more than the individual repairs. This is the sequence that holds in almost every case.
Why fix data loss first?
Sync failures and duplicate creation come first, always. Every other repair is built on the assumption that the data underneath is complete, and none of them survive if it is not.
Why agree definitions before configuring?
What a qualified lead is, when a stage is genuinely exited, which touch gets credit. These are written down before anything is configured, because encoding a disagreement produces a system both teams distrust.
Why is scoring third rather than first?
Now worth rebuilding, on data that arrives complete and definitions people accept. Doing this first is the most common sequencing mistake, and it is the one that produces a second rebuild a year later.
Why does adoption come last?
Layouts, training and a named person who can change a field. Last, because until the previous three are sorted, you are training people to use a system that is still wrong.
Or tell me
Describe it in three lines.
Whatever you ticked above travels with the message, so the reply starts from your symptoms rather than from "tell me about your business". If the honest answer is that you do not need the work, that is what the reply will say.
Serhii Skrypnyk · RevOps Architect · 7 Salesforce certifications · on the platform since 2018. Answered within one working day
Measure it before you argue about it.
The free readiness check runs in your browser on four report exports and produces a verdict without any org access. If it says the foundation holds, you have saved yourself a project.
Serhii Skrypnyk · RevOps Architect · 7 Salesforce certifications · on the platform since 2018
Questions
Asked often enough to answer here.
Why does sales ignore the leads marketing sends?
Usually a scoring model tuned for a buying process that has since changed, awarding points for behaviour that no longer indicates intent. Top-scoring prospects accumulate value from years-old activity, so the list is ordered by history rather than readiness. Adding rules makes it worse; the model must be back-tested against deals that actually closed.
Why do records go missing between Pardot and Salesforce?
A sync error queue with no owner. Lead-to-contact conversion drops records and prospects fail to arrive, but nothing alerts anyone. The first useful step is counting how many records are affected. Open the sync error queue and read the oldest entry's date: if it predates this quarter, nobody is watching it.
Why do marketing and finance report different pipeline numbers?
It is an attribution and data model problem rather than a reporting one. The systems disagree on what a qualified lead is, when revenue is recognised and which touch gets credit. Building more dashboards cannot resolve a definition that was never agreed.
Which Salesforce problem should be fixed first?
Anything losing records comes first, because every other repair assumes the data underneath is complete. Then definitions nobody agreed, then scoring and automation, then adoption and ownership. Rebuilding scoring first is the most common sequencing mistake and produces a second rebuild a year later.
Is it worth paying for a diagnosis before a repair?
Three of the six common symptoms look identical from the outside and are not the same problem. A €500 check measures four things in three days, read-only, and is credited in full against any engagement booked within sixty days. Roughly a third end with a recommendation that costs nothing.