Introducing MarketingSoda Refine: 7-Dimension Data Quality Scoring for HubSpot

Your HubSpot database is decaying right now. Not metaphorically — literally. Job titles are going stale at 65.8% per year. Email addresses are invalidating at 37.3%. Phone numbers are disconnecting at 42.9%. By the time you read this sentence, several records in your CRM have become materially less accurate than they were yesterday.

You already know this intuitively. You have seen the campaigns that underperform, the sequences that go unanswered, the lead routing that breaks silently. What you have not had — until today — is a way to measure it.

Today we are launching MarketingSoda Refine: the first purpose-built, multi-dimensional data quality scoring system designed natively for HubSpot. Refine scores every contact in your database across seven dimensions, assigns an A-F grade, and gives you the tools to act on what it finds — automatically.

This is not another enrichment tool. It is not a duplicate finder with a dashboard bolted on. Refine is a complete data quality operating system for HubSpot teams who are tired of building campaigns on data they cannot trust.

The Problem: You Cannot Fix What You Cannot Measure

We have spent the past year talking to RevOps teams, marketing operations leaders, and HubSpot admins about data quality. The pattern is remarkably consistent.

Every team knows their data has problems. Almost no team can quantify those problems with precision. They know their bounce rate is creeping up. They know certain segments underperform. They suspect duplicates are inflating their counts. But when asked "what percentage of your database is actually campaign-ready?" — the honest answer is usually "we don't know."

This is not a knowledge gap. It is a tooling gap.

The tools available to HubSpot teams today offer binary checks — is this email valid or not, is this a duplicate or not — or single-dimension analysis. No tool provides a composite quality score across multiple dimensions. No tool tells you that a contact has a valid email address but a job title that is 14 months stale, a phone number that fails format validation, and a company size field that contradicts the industry classification. Those compound quality issues are invisible to every tool in the current ecosystem, and they are exactly the issues that cause your campaigns to underperform in ways that are difficult to diagnose.

We wrote about the financial impact of this in The Real Cost of Bad HubSpot Data. The IBM estimate is $3.1 trillion annually across the US economy. The per-organization number is $15 million. But those aggregate figures obscure the more practical reality: bad data costs you in slow, invisible ways that never surface as "data quality problems" in your retrospectives. They surface as "our messaging needs work" or "sales isn't following up fast enough."

Refine makes the invisible visible.

The Refine Scoring Model: 7 Dimensions, One Grade

The core of Refine is a scoring framework that evaluates every contact in your HubSpot database across seven distinct quality dimensions. Each dimension is scored independently, weighted according to your configuration, and combined into a single composite score that maps to an A-F letter grade.

Think of it as a credit score for your CRM data. A single number that tells you whether this record is trustworthy enough to act on — and if not, exactly why.

The 7 Dimensions

Grade Thresholds

Freshness Scoring in Detail

The freshness dimension deserves specific explanation because it is the most commonly underestimated quality factor. Refine uses time-decay brackets based on the last verified date for each record:

A contact that was enriched last week scores 100 on freshness. A contact that has not been touched in 14 months scores 20. When freshness carries a 15% weight in the composite score, that difference alone can move a record from a B to a C — which is the threshold between "good for personalized outreach" and "only suitable for broad campaigns."

Most databases have a large population in the 365+ day bracket. We detailed the field-by-field decay rates in Building a Data Decay Strategy. The short version: if you are not actively re-enriching your database on a rolling basis, your freshness scores are degrading every day.

Customizable Scoring Profiles

The default weights above represent a balanced starting point. But different teams weight quality differently based on their GTM motion, and Refine accounts for this.

An outbound SDR team cares intensely about accuracy and freshness — they need correct names, current job titles, and working phone numbers. Completeness matters less if the core contact fields are populated. Their scoring profile might weight accuracy at 30% and freshness at 25%, with completeness dropped to 10%.

A marketing team running large-scale email campaigns cares most about validity and uniqueness — they need deliverable email addresses and deduplicated lists. Their scoring profile might weight validity at 25% and uniqueness at 20%.

An ABM team targeting enterprise accounts cares about enrichment coverage and consistency — they need deep firmographic data and records where the fields tell a coherent story. Their profile might weight enrichment coverage at 20% and consistency at 25%.

Refine ships with three pre-built scoring profiles — Outbound, Marketing, and ABM — and allows you to create custom profiles with any weight distribution that adds to 100%. Each profile produces its own letter grade for every contact, so you can evaluate the same database through different quality lenses depending on the motion you are planning.

How Refine Works: From Connection to Action

Getting started with Refine takes less than five minutes.

Step 1: Connect your HubSpot account. One-click OAuth. No API keys to configure, no custom integrations to maintain. Refine connects as a native HubSpot integration with read and write permissions scoped to contact and company properties.

Step 2: Automatic scoring begins. Refine scores your entire contact database on first connection. For a 50,000-contact database, initial scoring completes in under an hour. Scores are written back to HubSpot as custom contact properties, so they are immediately available in your existing lists, workflows, and reports.

Step 3: Review your dashboard. The Refine dashboard shows your database-wide grade distribution — what percentage of contacts scored A through F — along with dimension-level breakdowns, trend analysis over time, and the ability to drill into any segment or individual record.

Step 4: Set up quality gates. Quality gates are automated rules that use Refine scores to control campaign enrollment, sequence inclusion, and lead routing. You define the threshold — for example, "only enroll contacts with a B or above in outbound sequences" — and Refine enforces it continuously.

Step 5: Activate remediation. For records that score below your threshold, Refine can automatically trigger enrichment workflows, flag records for manual review, or quarantine contacts until their quality improves.

The scoring is not a one-time snapshot. Refine re-scores your database on a rolling basis, catching decay as it happens and updating grades in real time. When a contact's freshness decays from a 90-day bracket to a 180-day bracket, their score updates and your quality gates respond automatically.

Beyond Scoring: The Full Refine Platform

Data quality scoring is the foundation of Refine, but it is not the entire product. Scoring tells you what is wrong. The rest of the platform helps you fix it — and prevent it from happening again.

Waterfall Enrichment

Refine includes a multi-provider enrichment engine that cascades through up to six data providers to maximize field coverage. Unlike single-source enrichment tools that cap out at ~40% coverage, Refine orchestrates across Apollo, Cognism, ZoomInfo, Hunter.io, and other providers to reach 80-95% field coverage.

Refine adds field-level survivorship logic on top of the waterfall — when two providers return different values for the same field, Refine selects the most reliable value based on provider confidence scores and field provenance tracking. Every enrichment decision is auditable: you can see which provider supplied which field and when.

Probabilistic Deduplication

Refine uses Splink, an open-source probabilistic record linkage library, to identify duplicates that exact-match tools miss entirely. Same person, different email address. Same company, slightly different name. Same contact entered by two different reps with minor spelling variations.

HubSpot's native dedup tool catches less than 20% of actual duplicates in a typical database. We documented this gap in How to Deduplicate Your HubSpot Contacts. Refine's probabilistic matching catches the other 80% — and it runs both as a batch process for your existing database and as a real-time prevention layer for new records entering HubSpot.

Data Standardization

Phone numbers in 14 different formats. Job titles with hundreds of variations that mean the same thing. Company names with inconsistent capitalization, abbreviations, and legal suffixes. Standardization problems are the silent saboteur of segmentation logic and reporting accuracy.

Refine normalizes automatically:

• Phone numbers to E.164 international format
• Job titles to a canonical taxonomy with seniority mapping
• Company names with consistent formatting, abbreviation expansion, and legal suffix handling
• Country and state names to ISO standard codes

Validation

Beyond format standardization, Refine validates that field values are actually correct:

• Email verification — deliverability checks that catch invalid, disposable, and role-based addresses before they damage your sender reputation
• Phone validation — confirms number assignment and line type (mobile, landline, VoIP)
• Address verification — postal address standardization and deliverability confirmation

Decay Monitoring

Refine tracks freshness at the field level, not just the record level. It knows that a contact's email was verified 30 days ago but their job title was last confirmed 11 months ago. When individual fields cross staleness thresholds, Refine can automatically trigger targeted re-enrichment — refreshing only the fields that need it, rather than burning credits on a full re-enrichment of fields that are still current.

Quality Gates and Automation

Quality gates are where scoring translates into operational impact. Refine integrates with HubSpot Workflows to enforce data quality standards automatically:

• Campaign gates: Only send to contacts that meet a minimum quality threshold
• Sequence gates: Only enroll contacts in sales sequences if their accuracy and freshness scores pass
• Routing gates: Only route leads to reps if critical fields are populated and verified
• Enrichment triggers: Automatically enrich records that fall below a quality threshold
• Quarantine rules: Isolate records with F grades from all active campaigns until remediation is complete

These gates run continuously. As scores change — due to decay, enrichment, or manual updates — enrollment decisions update in real time. Your campaigns never run on data that fails your quality standards, because the gate catches it before it enters the workflow.

What Makes Refine Different

The HubSpot ecosystem has enrichment tools, dedup tools, and validation tools. What it has never had is a tool that measures data quality comprehensively, assigns a grade you can act on, and then provides the full remediation stack to improve that grade over time.

We wrote a detailed comparison of Breeze Intelligence's capabilities and limitations in HubSpot Breeze Intelligence: An Honest Review. Breeze is a reasonable starting point for teams that need basic enrichment inside HubSpot's native ecosystem. Refine is for teams that need to understand and control their data quality at a deeper level — and who want the flexibility to bring their own enrichment providers rather than being locked into a single source.

Pricing

Refine is priced by database size, not by credits that expire or enrichment calls that get marked up.

All plans include the complete scoring engine. No credit systems, no per-record fees for scoring, no expiring allocations. Your contacts are scored continuously as part of your subscription.

Enrichment costs are transparent — Refine does not mark up per-record data access. The platform charges for orchestration and intelligence, not for data.

Try It Before You Buy It: The Free Audit

You do not need to sign up, enter payment information, or connect your HubSpot account to see what Refine can do. We built a free data quality audit that analyzes a sample of your HubSpot database and gives you a preview of your grade distribution across all seven dimensions.

The audit takes about three minutes. You will get:

• Your overall database quality grade
• Dimension-level scores showing where your data is strongest and weakest
• An estimate of what percentage of your database is campaign-ready
• Specific recommendations for the highest-impact remediation actions

We have run this audit for hundreds of HubSpot teams during our beta period. The results are consistently eye-opening — most teams overestimate their data quality by 20-30 percentage points. The gap between perceived quality and measured quality is where campaigns go to underperform.

Run the free audit now and see where your HubSpot data actually stands.

What Happens Next

Refine is available starting today. Connect your HubSpot account, run your first score, and start making data quality decisions based on measurement rather than intuition.

If you are not ready to connect but want to stay informed, join the waitlist at marketingsoda.ai for product updates, early access to new features, and data quality content from our team.

Over the coming weeks, we will be publishing deep dives into each component of the Refine platform:

• Why We Built Refine — the backstory and the gap we set out to close
• The Data Standardization Problem — why normalization is the unsexy foundation of everything else
• How scoring profiles work in practice, with real configuration examples
• Quality gate recipes for the most common HubSpot workflow patterns

Your data is either an asset or a liability. Today, for the first time, you can measure which one it is — across every dimension that matters — and take automated action to shift the balance.

Start your free audit