Cleanup was a project, not a process
A one-off merge run fixes the backlog and changes nothing upstream. Without scheduled jobs and a review routine, the same duplicates and gaps come back at the rate your org creates records.
Salesforce · Cloudingo Data Quality Consulting
Cloudingo Salesforce data quality consulting that holds after the cleanup
Salesforce data quality consulting decides what good CRM data means for your business, who owns it and how it is checked, then puts a tool behind those decisions. Twopir Consulting builds that operating model with your admins, ops teams and data owners, and uses Cloudingo to enforce it: your matching policy becomes filters, your survivorship rules become merge rules, and your review routine becomes scheduled jobs. Standards first, tool second, measured every month.
12+
500+
250+
40+
Trusted by 500+ organizations — including software, legal, real estate and healthcare teams who rely on their Salesforce data being right.
What We Govern
- Salesforce Partner
- Data Standards
- Data Ownership
- Matching Policy
- Survivorship Rules
- Import Hygiene
- Data Quality Scorecards
- Cloudingo Optimization
The cleanup worked. Then the data drifted back.
Most Salesforce orgs have had at least one big cleanup. Six months later the duplicates are back, the picklists have drifted and the sales team has stopped trusting the reports again. A tool did not fail there; nobody decided how the data should stay clean once the project ended. These are the six gaps we find most often.
No one owns any object
Sales assumes ops owns accounts, ops assumes the admin does, and the admin only hears about it when a report breaks. Data with no named owner has no one to approve a rule or act on a finding.
Standards live in people's heads
Is it "Inc." or "Incorporated", "CA" or "California", and which fields must a lead have before it is routed? If the answer is not written down, every import and every new hire sets a different standard.
Merge rules nobody can explain
Someone configured the master-record and field rules years ago. A merge keeps the wrong phone number or owner, a rep complains, and automation gets switched off rather than fixed.
Entry points bypass everything
List imports, web forms, integrations and marketing sync create records faster than any cleanup removes them. If new data does not pass the same checks as old data, the backlog never shrinks.
Nobody measures it
"The data is bad" is a complaint, not a metric. Without a baseline per object and a trend line, data quality cannot be prioritized against other Salesforce work, so it never is.
Definition
Data quality consulting, defined
Salesforce data quality consulting is the work of deciding what accurate, complete and unique CRM data means for your business, who is accountable for it, and how it is kept that way, then configuring tools and routines to enforce those decisions. With Cloudingo, that means turning a written matching policy into filters, a survivorship policy into merge rules, an import standard into import checks, and a review routine into scheduled automation jobs and a monthly scorecard.
Three parties are involved, and it helps to keep them apart. Cloudingo, from Symphonic Source, is a cloud application that connects to Salesforce over the API through an OAuth connection and finds, merges, updates and imports records. Salesforce holds the records and runs your own validation rules, flows and duplicate rules. Twopir Consulting does the consulting around both: the standards, ownership, policy, configuration and ongoing tuning described on this page. Product behavior is documented in Cloudingo's help center.
Why Cloudingo rather than Salesforce's native tools alone? Cloudingo's own comparison notes that Salesforce's native merge handles three records at a time and that native matching compares a limited set of fields, while Cloudingo merges larger groups and adds field-level merge rules and scheduled automation. Neither tool decides what a duplicate is for your business, which record should survive, or who signs off on a rule. That is the part this service covers.
New to the product itself? Our introduction to Cloudingo for Salesforce walks through what it does and how a first deduplication run works.
Eight decisions a data quality program makes
CRM data governance does not need a committee or a hundred-page policy. It needs these eight decisions made on purpose, written down where an admin can find them, and reflected in how Cloudingo is configured.
| Decision | What we decide with you | What goes wrong when it is skipped |
|---|---|---|
| Ownership | A named data owner for each core object (leads, contacts, accounts and any custom object that matters), and the stewards who review findings and approve rule changes. | Findings sit in a dashboard nobody opens, and every rule change becomes a debate between teams. |
| Data standards | Required fields by record stage, picklist values, naming conventions for companies, countries, states and phone numbers, and what "complete" means for a routed lead. | Each import and each team applies its own format, and standardization work starts again every quarter. |
| Matching policy | What counts as a duplicate on each object and across objects (lead to contact, lead to account), which fields decide it, and where exact matching ends and fuzzy matching begins. | Filters that are either too loose and merge different customers, or too tight and miss the real duplicates. |
| Survivorship | Which record survives a merge (newest, oldest, most complete, most activity), which value wins for each important field, which fields travel together, and who owns the result. | Merges that keep a stale phone number or an inactive owner, and users who stop trusting automation. |
| Entry points | Which sources create records (imports, web forms, integrations, Marketo) and which checks each one passes through before a record lands. | New duplicates arrive faster than scheduled jobs remove them. |
| Automation vs review | Which duplicate groups merge automatically on a schedule, which wait for a person, and which records are exceptions that must never auto-merge. | Either nothing is automated and the backlog grows, or everything is and a bad merge reaches a key account. |
| Measurement | The handful of metrics per object that define data quality for you, the baseline, the target, and the review cadence. | No way to show the work is paying off, so data quality loses every prioritization meeting. |
| Change control | Who may create or edit filters, merge rules and automation jobs, how changes are tested and recorded, and how rules are shared between connected environments. | Rules edited in place by whoever has a login, with no record of why last month's merges behaved differently. |
Who owns Salesforce data quality, role by role
Data quality is a shared job, and shared jobs fail when nobody knows their part. This is the ownership model we set up in most engagements, scaled to the size of your team. In a smaller company one person may hold two of these roles; the responsibilities still need a name against them.
Executive Sponsor
Usually a revenue, operations or IT leader who needs the data to be right and can make it a priority.
- Approves the data quality policy
- Settles disputes between teams
- Sees the scorecard every quarter
Data Owners
One per core object, from the team that depends on it most: sales ops for accounts, marketing ops for leads.
- Own the standards for their object
- Approve matching and survivorship rules
- Accept or reject exception records
Data Stewards
The people closest to the records, who work the review queues and notice when a standard stops fitting.
- Review merges held for a person
- Work unmerged groups each month
- Raise rule changes with evidence
Salesforce Admin
Keeps Cloudingo and Salesforce aligned, so a rule that works in the tool does not collide with the org.
- Maintains filters, rules and jobs
- Manages the connected user's access
- Checks flows and validations before merges
Source Owners
Whoever runs a channel that creates records: list imports, web forms, an integration or marketing sync.
- Route imports through agreed checks
- Map source fields to the standards
- Fix the source, not just the record
Twopir Consulting
The architect and facilitator: we design the model with you, configure it, and stay to tune it if you want us to.
- Drafts the policy for owners to approve
- Configures Cloudingo to match it
- Runs the monthly review with your team
Where Cloudingo enforces the policy
Cloudingo is the enabling tool, not the strategy. This is how we map the policy onto it, and what our consulting adds on top of each feature. Some capabilities depend on the Cloudingo edition, so we check licensing against the policy before anything is configured.
| Governance need | What Cloudingo provides | What our consulting adds |
|---|---|---|
| Know where you stand | A dashboard of duplicate counts and recent jobs, field analysis showing the number, type and usage of fields on an object, and filters that work like reports and change nothing until a merge is submitted. | A baseline per object against your own standards, so the first scorecard has a starting point the owners agree with. |
| Enforce the matching policy | Filters built from templates or a drag-and-drop editor, with exact, fuzzy and synonym matching, and cross-matching filters that compare two different fields, such as email against alternate email. | Filters written from the policy, started tight and loosened deliberately, with each one tied to the decision it enforces. |
| Enforce survivorship | Merge rules that choose the master record and the winning value per field, field groups that keep related values (an address block) together, and owner mapping for inactive owners and queues. | Rules that match what each data owner signed off, and exceptions for the records that must never merge automatically. |
| Keep standards in place | Mass update, transform and standardize, and find and replace across records; U.S. mailing-address validation is an add-on on every plan. | The standards themselves, and a schedule for applying them, so standardization is routine rather than a quarterly fire drill. |
| Protect the entry points | An import wizard that checks CSV rows against existing records before insert; real-time merge on Professional and Enterprise; API integration on Enterprise for incoming data. | A rule for each source on which route it takes, and the import templates your teams actually use. |
| Run it without reminders | Automation jobs scheduled daily, weekly or on demand; automated merges that run a whole filter without a user logged in. | A job calendar that matches your record volume and your review capacity, and a check that jobs still run after org changes. |
| Recover from a bad merge | Undo & Restore on Professional and Enterprise, which unmerges records merged in Cloudingo and reattaches related data from backups kept up to 45 days on Cloudingo storage. Cloudingo states it is not a substitute for a backup solution. | A recovery procedure your stewards know before they need it, and a reminder that your Salesforce backup plan still applies. |
| Control who changes what | Permission-based logins and roles per feature, and Cloudingo Intelligence, which monitors logins, permissions, installed apps, API usage and system changes and sends alerts. | Role assignments that follow the change-control decision, so the people who approve rules are not the only people who can break them. |
Cloudingo is priced per Salesforce org and billed annually rather than per user, and seats, custom-object merging, real-time merge and API integration differ by edition. Current details are on Cloudingo's pricing page. We confirm which edition your policy needs before you renew or upgrade, not after.
From assessment to roadmap to a running program
A consulting engagement starts with evidence and ends with a routine your team runs. Each phase has a deliverable, and the specialist work in the middle, such as deduplication or a cleansing program, is scoped from the roadmap rather than guessed at the start.
Assess
We measure what is in the org today: duplicates by object, field completeness, standards drift and where new records come from. When you need it as a standalone diagnostic, it is our Salesforce data quality audit.
DeliverableBaseline and prioritized findings
Define
The eight governance decisions, drafted by us and approved by your owners: standards, ownership, entry-point rules, measurement and change control.
DeliverableData quality policy and roadmap
Design
Matching and survivorship policy per object and across objects, tested as filters that report before anything merges. The merge work itself runs as Cloudingo deduplication services.
DeliverableApproved matching and merge policy
Configure
Cloudingo set up to the policy: connection, permissions, filters, merge rules, import templates and automation jobs. A first-time setup is delivered as Cloudingo implementation services.
DeliverableConfiguration mapped to each decision
Remediate
The backlog worked down in a planned order, object by object, with stewards reviewing what automation should not decide. Standardization, stale records and import hygiene run as Cloudingo data cleansing services.
DeliverableBacklog cleared against the baseline
Operate & Optimize
Scheduled jobs, a monthly scorecard review with your owners, and rule tuning as your org, sources and team change. Your admins can run it alone, or we stay on.
DeliverableRunbook, scorecard and review cadence
How we measure data quality month after month
Cloudingo does not produce a data quality score, so the scorecard is ours: six dimensions, measured per object with Cloudingo's filters and field analysis alongside Salesforce reports. The targets are set with your owners from the baseline, not borrowed from a benchmark.
| Dimension | What we track, and with what |
|---|---|
| Uniqueness | Duplicate groups per object and across leads and contacts, from the same Cloudingo filters that enforce the matching policy. |
| Completeness | Fill rate of the fields your standards require at each record stage, from field analysis and Salesforce reports. |
| Consistency | Values that break a standard: company-name variants, country and state formats, picklist drift. |
| Validity | Values in the wrong shape for their field, and U.S. mailing addresses checked where address validation is licensed. |
| Freshness | Records untouched for longer than your agreed window, and test or inactive records still counting against storage. |
| Ownership | Records owned by inactive users or queues, and review queues older than the agreed turnaround. |
Six situations that call for a data quality program
Teams rarely call us to say "we need governance". They call because something specific broke. These are the starting points we see most, and what the first move is in each.
You Have Just Bought Cloudingo
The license is signed and the first instinct is to run every template filter and merge. That is the moment to slow down.
- Signal
- A new license, no rules yet, a large backlog
- First move
- Matching and survivorship policy before the first merge
- Then
- Configuration to that policy, and a baseline to measure from
Cloudingo Runs, Nobody Trusts It
Cloudingo has been in place for years. Automation was paused after a bad merge and nobody remembers why the rules look the way they do.
- Signal
- Paused jobs, manual merges, complaints about lost values
- First move
- Review every filter and merge rule against a written policy
- Then
- Rules re-approved by owners, automation switched back on
After a Migration or Org Merge
Data arrived from another CRM, or two Salesforce orgs became one, and nobody cleaned it on the way in.
- Signal
- Duplicates across objects, mixed standards from two sources
- First move
- A baseline, then a cleanup order agreed object by object
- Then
- One set of standards, applied to every source going forward
Marketing and Sales Disagree
Leads that are already contacts, accounts nobody can route, and two teams each sure the other created the mess.
- Signal
- Lead-to-contact duplicates, routing and attribution disputes
- First move
- Ownership of leads and accounts, and a cross-object matching rule
- Then
- Entry-point checks on forms, imports and Marketo data
Getting Ready for AI and Automation
Forecasts, routing, Flow automation and AI features all act on the records you have. Duplicates and gaps become wrong answers at speed.
- Signal
- An AI, analytics or automation project with a data dependency
- First move
- Measure the objects that project reads, before it goes live
- Then
- Standards and monitoring on those objects as a launch condition
Data Quality Is a Side Job
Your admin cares about the data and has no time for it. Jobs drift, the review never happens, and nothing tells you it has slipped.
- Signal
- No scorecard, no review, one stretched admin
- First move
- A light policy and a monthly review your admin can sustain
- Then
- Ongoing optimization with us, or a clean handover
Already standardized on another tool? We also support teams running Validity DemandTools and Duplicate Check for Salesforce. The governance model on this page applies whichever tool enforces it.
What a routine looks like in practice
These two stories are published by Cloudingo, not by Twopir, and we were not involved in either. They are here because they show the pattern this page argues for: a defined cleanup order, filters run on a schedule and a regular human review. Every figure is Cloudingo's, as published.
Docker, Inc.
Cloudingo reports that an app creating an account per credit-card transaction left Docker with as many as sixty accounts for one company, and that Docker worked accounts first, then leads, then contacts, then leads against contacts.
60 Accounts for one company
$50K–$1M Yearly cost of bad data, as reported
4 Steps in the cleanup order
Hotze Health & Wellness Center
Cloudingo reports that Hotze migrated from Microsoft Dynamics without cleaning its data, then made Cloudingo a daily routine with a monthly review of the records left unmerged.
~30K Patients on the roster, approaching
~12 Filters in the daily run
1 Review of unmerged records a month
A data quality partner who knows the whole org
Data quality sits where sales, marketing, operations and IT meet. Someone who only knows the tool can configure it correctly and still leave you with rules that fight the rest of your Salesforce org.
We start from your business, not the filter list
What a duplicate is, and which value should win, depends on how you sell, route and report. We write that down with your owners before a single rule is built, so the configuration has a reason behind every setting.
We are Salesforce architects first
Cloudingo notes that merges can be affected by Salesforce limits, triggers, flows and validation rules, and that it cannot troubleshoot those for you. We can, because we review your automation and sharing model as part of the work.
We fix the sources, not just the records
Imports, web forms, integrations and marketing sync are where most bad data starts. We trace problems back to the source and change it there, which is the only way the backlog stays small.
We are clear about what is ours
Cloudingo is licensed from Symphonic Source, the company that builds it. Twopir Consulting provides the strategy, configuration and ongoing work around it, and we say plainly when a native Salesforce feature or a different tool fits better.
We hand over what we build
You get a written policy, a configuration mapped to it and a runbook your admins can follow. When you want us to stay, ongoing Salesforce support covers the monthly review and rule tuning.
Data quality consulting questions, answered plainly
A data quality consultant decides, with your team, what good data means in your Salesforce org and then configures Cloudingo to enforce it. That covers data standards, ownership by object, a matching policy and a survivorship policy, the checks each entry point passes through, a measurement scorecard and change control. Those decisions become Cloudingo filters, merge rules, import templates and scheduled jobs, and a monthly review keeps them current as the org changes.
Cloudingo is a capable tool for finding, merging, updating and importing Salesforce records, but it does what it is configured to do. It does not decide what counts as a duplicate for your business, which record should survive a merge, who approves a rule or how quality is measured. Teams that buy the tool without those decisions usually get a good first cleanup and then watch the data drift back.
Each core object should have a named data owner from the team that depends on it most, supported by data stewards who review exceptions and a Salesforce admin who maintains the configuration. An executive sponsor approves the policy and settles disputes between teams. In a smaller company one person may hold several of these roles, but each responsibility still needs a name against it.
It depends on the policy. Cloudingo's Standard edition includes scheduled automation jobs and mass merge, update and delete on standard objects. Professional adds real-time merge, Undo & Restore, rapid import and merging on up to three custom objects, and Enterprise adds API integration and unlimited custom objects. Cloudingo is priced per Salesforce org and billed annually, and publishes current plan details on its pricing page. We confirm the edition against your policy before you renew or upgrade.
An audit measures where your data stands and produces prioritized findings; a cleanup works down the backlog. Data quality consulting includes both as phases and adds what keeps the result in place: standards, ownership, entry-point rules, measurement and a monthly review. If you only need the diagnostic, or only the deduplication or cleansing work, each is available as a separate service.
Cloudingo connects to Salesforce over the API with an OAuth connection and states that customer data is neither stored nor cached, except for the backups kept for Undo & Restore. Its pricing page lists SOC 2 Type II, GDPR and CCPA compliance and 256-bit SSL encryption, and Cloudingo describes itself as a Salesforce ISV that goes through periodic security reviews. Salesforce states that a security review is not a guarantee of an app's security, so we review the connected user's permissions with your security team as part of the work.
No. The governance model is the same whichever tool enforces it, and we also support teams running Validity DemandTools and Duplicate Check. If Salesforce's native duplicate and matching rules cover what your org needs, we will say so. When Cloudingo is the right fit, we configure it to your policy and help you get more from the license you already pay for.
Next Step
Show us the data, we will map the fix
A first conversation covers your core objects, where records come from, how Cloudingo is set up today (or whether you have it yet) and who owns the data: enough to say where a program should start and what the first phase would involve. No obligation to proceed.
Talk to a Salesforce architect who works with Cloudingo
