Somebody on the leadership team asks whether the CRM is healthy. An admin searches “Salesforce health check,” finds a built-in tool with that exact name, runs it, and comes back with a score in the eighties. Everyone relaxes. The score was about password policies and session timeouts. Nobody looked at a single deal, contact or account.

The name does a lot of damage. Each platform has some native tooling that touches data quality, and each covers a narrow slice of it. This paper lays the three side by side: what each tool checks, what it leaves out, which plan it needs, and how to run it. Then it covers the problems no native tool looks for, which are usually the ones distorting your forecast.

Native CRM health checks, side by side

Each platform’s native tooling answers a different question. Salesforce Health Check asks whether your security settings meet a baseline. Salesforce Optimizer asks whether your configuration is cluttered. HubSpot’s Data Quality Command Center asks whether records are duplicated or badly formatted. Pipedrive’s Merge Duplicates asks whether two people or organizations are the same. None asks whether the pipeline reflects reality.

ToolWhat it checksWhat it does not checkPlan or access neededTime to a first result
Salesforce Health CheckSecurity settings (session settings, password policies, network access) scored against Salesforce’s baseline or a custom oneAny record data: duplicates, missing fields, stale deals, ownershipSetup access (“View Setup and Configuration”)Minutes; it reads settings, not data
Salesforce OptimizerConfiguration: field counts and field usage, unused reports and dashboards, page layouts, validation and workflow rules, limitsWhether the values in your records are right, current or completeAdmin permissions to authorize and run itTrailhead says a scan usually finishes within an hour
Salesforce Duplicate ManagementDuplicates at entry, through matching rules and duplicate rules that alert or blockDuplicates already in the database (that takes duplicate jobs, which depend on edition), and everything that is not a duplicateVaries by edition; check yoursRules work from the moment they are active
HubSpot Data Quality Command CenterDuplicates, formatting issues on properties, property insights, unused workflows, a weekly digestPipeline hygiene, close dates, deal stages, ownership gaps, records missing the fields your reports needListed across most hubs at Starter and up; formatting automation needs Data Hub Professional or EnterpriseOpens immediately; the time goes into working the queue
HubSpot duplicate managementContact and company pairs HubSpot’s matching flags as likely duplicatesDuplicates its matching cannot see, such as the same person under two emails with different namesProfessional or Enterprise; custom duplicate rules need Data Hub Professional or EnterpriseOpens immediately; review is pair by pair
Pipedrive Merge DuplicatesPeople with a matching name plus a matching phone, email or organization; organizations it flags as likely matchesEverything else: missing fields, stale deals, ownership, pipeline accuracyAdmins, or users given the permissionOpens immediately; merging is manual

Sources: each vendor’s own documentation: Salesforce Help and Trailhead for Health Check, Optimizer and duplicate management, HubSpot’s Knowledge Base for the Data Quality Command Center and duplicate management, and Pipedrive’s Knowledge Base for Merge Duplicates and how it identifies duplicates. Plans and limits change often, so check the vendor page against your own subscription.

Is the Salesforce Health Check a data quality check?

No. Salesforce Health Check is a security tool. Per Salesforce’s Health Check documentation, it scores your org’s security settings, such as password policies, session settings and network access, against the Salesforce Baseline Standard or a custom baseline you choose. It does not open a single record. A perfect score tells you nothing about duplicates, empty fields or stale deals.

That makes it a useful tool with a misleading name. Run it, because weak session and password settings are worth knowing about. Then file the result under security, not under data.

The Salesforce tool closer to a health check of the org itself is Optimizer. Trailhead’s Optimizer module describes a report that lists findings by status, with the urgent ones marked for review, covering things like objects with too many fields, fields that hold no data, and unused reports and dashboards. It is a clutter check on configuration. Clutter matters, and we covered why in how to find unused fields in Salesforce and HubSpot, but a tidy configuration can still hold wrong data.

Duplicates in Salesforce sit with Duplicate Management: matching rules define what counts as a duplicate, and duplicate rules decide whether a user is warned or blocked on save. That protects new entries. It does nothing for the duplicates already in the database unless you also run duplicate jobs, and those depend on your edition.

What HubSpot’s Data Quality Command Center catches

HubSpot’s Data Quality Command Center is the closest any of the three comes to a native data health check. Per HubSpot’s knowledge base, it shows duplicate and formatting issues over a date range, recommended actions on specific properties, property insights, and unused workflows, plus a weekly data quality digest by email. It is a solid starting point for record-level hygiene.

What it covers is the record. What it leaves out is the business meaning of the record. It can tell you a phone number is badly formatted. It cannot tell you that a deal in “Contract sent” has had no buyer activity in two months, that half your open pipeline belongs to a rep who left, or that the lifecycle stage on a contact contradicts the deal attached to it. Those are the problems that make a forecast wrong.

Duplicates get their own tool. HubSpot’s duplicate management needs a Professional or Enterprise subscription, compares contacts on properties such as name, email, phone and company name, and has you review likely pairs. Custom duplicate rules need Data Hub Professional or Enterprise. The pairs it finds are real, but the matching only sees what the properties say, so the same person under two emails and a nickname can slip through. We covered why duplicates keep coming back in why HubSpot duplicates keep coming back, and the integration version of the problem in why HubSpot and Salesforce sync creates duplicates.

How does Pipedrive handle data quality natively?

Pipedrive’s native data quality tooling is mostly Merge Duplicates. Per Pipedrive’s support article on how Merge Duplicates identifies duplicates, it flags people who share a name plus a phone number, email or organization, using the same rules as its import tool, and lists possible duplicate organizations alongside them. Merging is limited to admins or users with the permission, and the record you choose as primary wins any conflict.

That is useful and narrow. The match leans on the name, so the same person entered with a misspelled name and a new email does not show up as a pair. Merging is one decision at a time. And beyond duplicates, Pipedrive’s native tools leave the rest of data quality to reports you build yourself and to marketplace apps.

The way duplicates arrive is a good reason to run it anyway. Integrations create them on their own: HubSpot’s knowledge base says that when a HubSpot contact is first created as a Salesforce lead with no account, creating an account or converting the lead to a contact in Salesforce creates a duplicate company in HubSpot. That example is HubSpot and Salesforce, but the mechanism, tools writing records with no shared key, applies to any CRM.

How do you run each native check?

Each native tool takes minutes to open, and the real work is the review afterwards. Run all of the ones your platform offers before you buy anything, because they are free with your subscription and they clear the obvious problems. Budget more time for the review than for the scan: deciding what to merge, fix or ignore is the slow part.

  1. Salesforce Health Check: Setup, search “Health Check” in Quick Find, open it, and read the high-risk settings first. Hand the result to whoever owns security.
  2. Salesforce Optimizer: Setup, search “Optimizer,” allow access, then run it and work the “Review Required” findings first, per Trailhead’s guide.
  3. Salesforce Duplicate Management: Setup, Duplicate Management, check which matching and duplicate rules are active on leads, contacts and accounts, and whether they alert or block.
  4. HubSpot Data Quality Command Center: Data Management, then Data Quality. Start with duplicates, then formatting issues on the properties your reports depend on, and switch on the weekly digest.
  5. Pipedrive Merge Duplicates: Tools and apps, Merge Duplicates. Work the list with whoever knows the accounts, since the primary record’s values win.

Write down what each tool found. The count of duplicates and formatting issues is your baseline, and it tells you how fast new problems arrive once you check again in a month.

Each native tool takes minutes to open, and the real work is the review afterwards.

01Run the native scanMinutes per tool,included in yoursubscription.02Decide what tomerge or ignoreBudget more time herethan for the scan, withwhoever knows theaccounts.03Write down thecountsDuplicates andformatting issues areyour baseline.04Check again in amonthThe change in the countsshows how fast newproblems arrive.
01Run the native scanMinutes per tool, included in yoursubscription.02Decide what to merge or ignoreBudget more time here than for the scan,with whoever knows the accounts.03Write down the countsDuplicates and formatting issues are yourbaseline.04Check again in a monthThe change in the counts shows how fastnew problems arrive.

Illustrative process · no data

Each native tool takes minutes to open, and the real work is the review afterwards.
StepDescription
Run the native scanMinutes per tool, included in your subscription.
Decide what to merge or ignoreBudget more time here than for the scan, with whoever knows the accounts.
Write down the countsDuplicates and formatting issues are your baseline.
Check again in a monthThe change in the counts shows how fast new problems arrive.

What no native CRM health check covers

No native tool checks whether your CRM matches what is happening in deals. That means pipeline hygiene (deals past their close date, stages that have not moved, pushed close dates), ownership (records owned by people who left), cross-object integrity (contacts, companies and deals that contradict each other) and freshness (records nobody has touched or verified in months). Those gaps drive bad forecasts and bad reports.

Those gaps exist because they need judgment about how your team sells, which a platform-wide tool cannot know. A deal sitting in a stage for thirty days is normal in one business and a red flag in another. A missing industry field is harmless until a territory or routing rule depends on it.

Across the 150+ B2B companies we’ve worked inside, roughly 7 in 10 had material CRM data accuracy problems. The ones that move a forecast, such as close dates already in the past, owners who have left and stages that mean different things to different reps, tend to sit outside what the native checks look at, so add those three to the monthly recheck. If your HubSpot reports disagree with each other, why HubSpot reports don’t match covers the usual causes. For what “clean enough” means once AI tools start reading the data, see AI-ready CRM data.

When does an outside CRM diagnostic make sense?

An outside diagnostic makes sense once the native tools are clean and the numbers still do not add up: the forecast misses, reports disagree, or leadership has stopped trusting the dashboard. Native tools cover the record. A diagnostic covers how records relate to each other and to how your team sells. If you have not run the native tools yet, start there; they are free.

The free CRM Diagnostic from AeolusGTM connects read-only to HubSpot, Salesforce or Pipedrive and runs 40+ automated checks across data quality, pipeline hygiene, freshness, reachability and ICP fit. The score comes from deterministic rules, not an LLM, and the connection never writes to your CRM. You get a detailed scored report within a few hours of connecting, then we walk through it with you live. If you only want a quick read on a HubSpot portal first, the free HubSpot audit tool gives you a self-score.

Trust the name, or check the data

You can run the tool called “Health Check,” report the score upward and call the CRM healthy. The score will look good, because it never looked at the data. Or you can run the native tools for what they are, accept that they cover duplicates, formatting and configuration, and go find out what they leave out before the next forecast call finds it for you.

Start this week: open the native check your platform offers, run it on contacts or accounts, and write down two counts, duplicates and formatting issues. Put a reminder in the calendar to run it again in a month and compare the two counts.

The native tools stop at duplicates, formatting and configuration. AeolusGTM is CRM, RevOps and GTM advisory, with Atlas connecting to your CRM to find gaps the CRM can’t show natively. If you want to see what your native tools miss, book your free diagnostic.