Data Quality & Freshness

Clean, complete, and current data is the foundation of every reliable sales forecast, every accurate pipeline report, and every effective automation in SalesOS. The Data Quality & Freshness module gives you continuous visibility into the health of your CRM data, surfaces records that need attention, and provides actionable recommendations to maintain high standards without manual auditing.

What Data Quality Means in SalesOS

Data quality in SalesOS is measured across four dimensions that together determine how trustworthy and actionable your records are:

• Completeness: Are required and recommended fields populated? A lead without a phone number or a deal without a close date reduces your team's ability to act.
• Accuracy: Do field values conform to expected patterns? An email address without an "@" symbol, a phone number with too few digits, or a deal amount of $0 on an enterprise opportunity signals inaccuracy.
• Consistency: Are related records aligned? If a contact's company name does not match the linked account name, or if a deal's stage does not align with its probability, there is an inconsistency.
• Freshness: Has the record been updated recently enough to remain relevant? A lead untouched for 90 days or an account with last activity 6 months ago is likely stale.

These four dimensions combine into a single Data Quality Score for every record and roll up into org-wide, team-level, and per-rep views.

---

Data Quality Score

Every record in SalesOS receives a quality score from 0 to 100. The score updates in real time as fields are added, modified, or become stale.

How It Is Calculated

The score is a weighted average across the four dimensions:

Score Ranges

Field-Level Scoring

Drill into any record to see which fields contribute to or detract from its score. Each field shows:

• Whether it is populated (completeness contribution)
• Whether it passes validation (accuracy contribution)
• When it was last updated (freshness contribution)
• Any cross-record conflicts (consistency contribution)

This granularity helps reps know exactly what to fix rather than facing an abstract "low quality" warning.

---

Quality Dashboard

The quality dashboard provides an org-wide view of data health with trends over time. Access it from Analytics > Data Quality in the main navigation.

Dashboard Components

Overall Quality Score: A single number representing your organization's average data quality across all record types. Displayed as a large metric with a trend arrow showing improvement or degradation over the past 30 days.

Score Distribution Chart: A histogram showing how many records fall into each quality rating band (Excellent, Good, Fair, Poor). Healthy organizations have a right-skewed distribution with most records in the Good or Excellent range.

Trend Line: A 90-day trend of your overall quality score, broken down by dimension. Spot whether freshness is the primary driver of decline or if completeness gaps are growing.

Top Issues Table: The ten most impactful quality issues across your org, ranked by the number of affected records. Each row shows the issue type, affected record count, estimated impact on reporting accuracy, and a one-click link to the affected record list.

Quality by Record Type: A breakdown showing average quality scores for leads, contacts, accounts, and deals independently. Quickly identify which object type needs the most attention.

---

Freshness Indicators

Freshness measures whether records are being actively maintained. In a fast-moving sales environment, data becomes stale quickly, and stale data leads to missed follow-ups, incorrect forecasts, and poor customer experiences.

How Freshness Is Determined

Each record type has an expected update cadence:

Stale Record Identification

Records exceeding their staleness threshold are flagged with a freshness indicator:

• Fresh (green): Updated within the expected cadence.
• Aging (amber): Approaching the staleness threshold (within 25% of the limit).
• Stale (red): Exceeded the threshold and requires attention.

Last-Updated Timestamps

Every record displays a "Last Updated" timestamp on its detail page, along with what changed and who made the update. The quality system distinguishes between meaningful updates (field value changes, new activities logged) and trivial updates (viewing the record, automated sync touches).

Freshness Alerts

Configure alerts to notify record owners when their records cross freshness thresholds:

• In-app notification: A badge in the sidebar shows the count of stale records requiring attention.
• Email digest: A weekly email listing records that became stale in the past 7 days.
• Manager alerts: Managers receive aggregate freshness reports for their team.

---

Quality Rules

Quality rules define what "good" data looks like for your organization. SalesOS ships with sensible defaults, and you can customize or add rules to match your data standards.

Required Fields

Define which fields must be populated for each record type and stage:

• Lead required fields: First name, last name, email or phone, source, status.
• Contact required fields: First name, last name, email, linked account.
• Account required fields: Company name, industry, employee count range, website.
• Deal required fields: Deal name, amount, close date, stage, linked account.

Required field rules are stage-aware. A lead in the "New" stage may only require name and email, but moving it to "Qualified" requires phone, company, and budget range.

Format Validation

Pattern-based validation catches common data entry errors:

Pattern Matching and Custom Rules

Create custom validation rules using pattern matching:

• Naming conventions: Account names must not contain "test", "demo", or "sample" (catches test data in production).
• Value ranges: Deal amounts must be between $1,000 and $10,000,000 (catches typos like $100M).
• Relationship rules: Every deal must have at least one associated contact (catches orphaned deals).
• Duplicate field detection: Flag records where the company name field contains an email address (common copy-paste error).

---

Automated Suggestions

Rather than simply flagging problems, SalesOS generates actionable cleanup suggestions that reps can accept with a single click.

Cleanup Recommendations

The system generates specific, actionable recommendations:

• Fill missing fields: "This lead is missing a phone number. Based on the email domain, the likely company phone is (555) 123-4567." Accept to auto-populate.
• Fix formatting: "The phone number '5551234567' should be formatted as '(555) 123-4567'." Accept to reformat.
• Update stale values: "This account's employee count (50-100) was set 18 months ago. LinkedIn shows 250+ employees." Accept to update.
• Correct inconsistencies: "This contact lists 'Acme Corp' as their company but is linked to the 'Acme Corporation' account. Update contact field to match?" Accept to align.

Merge Suggestions

When the system detects potential duplicate records, it generates merge suggestions:

• Duplicate detection criteria: Matching email addresses, similar names + same company, matching phone numbers, or matching LinkedIn URLs.
• Merge preview: Shows both records side by side with a recommended "winner" for each field based on recency and completeness.
• One-click merge: Accept the suggested merge, or customize which fields to keep from each record before merging.
• Merge history: All merges are logged in the audit trail with the ability to unmerge if needed.

Suggestion Priority

Suggestions are ranked by impact:

1. Critical: Data issues affecting active deals in late stages (e.g., missing close date on a deal in negotiation).
2. High: Issues on records with upcoming activities or active sequences.
3. Medium: Incomplete records that are actively being worked.
4. Low: Stale records or minor formatting issues on inactive records.

---

Quality by Record Type

Different record types have different quality profiles and require different attention strategies.

Leads

Lead quality is heavily influenced by source. Web form leads often have high completeness (forms enforce required fields) but may have accuracy issues (fake emails, placeholder phone numbers). Manually entered leads may have low completeness if reps skip optional fields during fast data entry.

Key metrics: Average completeness by source, conversion rate correlation with quality score, percentage of leads with valid email addresses.

Contacts

Contact quality tends to degrade over time as people change roles, companies, and contact information. The primary freshness challenge is keeping job titles, phone numbers, and email addresses current.

Key metrics: Percentage of contacts with verified email (no bounces), average time since last activity, contacts without linked accounts.

Accounts

Account data quality directly impacts territory planning, segmentation, and reporting accuracy. Missing industry, employee count, or revenue data limits your ability to segment and prioritize.

Key metrics: Percentage of accounts with complete firmographic data, accounts with no linked contacts, accounts with no recent activity.

Deals

Deal quality has the most direct impact on forecast accuracy. Missing amounts, stale close dates, incorrect stages, and disconnected contacts all reduce forecast reliability.

Key metrics: Deals with close dates in the past, deals without next steps or activities, deals stuck in a stage beyond the average cycle time, deals without linked contacts.

---

Quality by Team and Rep

Managers need visibility into how data quality varies across their team to provide coaching and hold reps accountable for CRM hygiene.

Team Quality View

The team view shows:

• Average quality score per rep: Ranked from highest to lowest to identify who maintains clean data and who needs support.
• Score trend per rep: Is a rep's data quality improving or declining over the past 30/60/90 days?
• Outstanding suggestions per rep: How many cleanup suggestions are pending action for each rep?
• Quality by activity type: Are reps logging activities with sufficient detail? Are notes being captured?

Rep Scorecard

Each rep has a data quality scorecard visible on their profile:

• Overall quality score for their owned records
• Breakdown by dimension (completeness, accuracy, consistency, freshness)
• Comparison to team average
• Number of cleanup actions taken this month
• Stale record count and trend

Manager Actions

Managers can:

• Set minimum quality score targets for their team
• Create cleanup sprints (time-boxed periods where the team focuses on data quality)
• Assign specific cleanup tasks to reps based on outstanding suggestions
• Include data quality metrics in performance reviews and 1:1 discussions

---

Decay Detection

Data quality naturally decays over time without active maintenance. Decay detection identifies when scores are trending downward and alerts you before quality drops below acceptable thresholds.

How Decay Is Detected

The system tracks quality scores for every record over time and identifies:

• Individual record decay: A specific record's score dropping by 10+ points over 30 days.
• Cohort decay: A group of records (e.g., all leads from a specific campaign) collectively declining.
• Dimensional decay: A specific quality dimension (usually freshness) declining across the org while others remain stable.

Decay Alerts

Configure threshold-based alerts:

• Alert when org-wide quality score drops below 75.
• Alert when any team's average drops below 70.
• Alert when more than 20% of active deals have a quality score below 60.
• Alert when the weekly decay rate exceeds 2 points.

Decay Prevention

The system recommends preventive actions based on decay patterns:

• If freshness is the primary decay driver, increase activity logging reminders.
• If completeness is declining (often after bulk imports), add validation rules to import workflows.
• If accuracy is dropping, review and tighten format validation rules.

---

Scheduled Quality Reports

Automate quality reporting to maintain awareness without manual dashboard checks.

Report Types

Configuring Reports

Navigate to Settings > Data Quality > Reports to configure:

• Which reports are enabled
• Delivery schedule (day of week, time)
• Recipients (can include users outside the platform via email)
• Thresholds for triggering alert reports (only send if score drops below X)

---

Improvement Workflows

Identifying quality issues is only valuable if it leads to action. SalesOS provides structured workflows for turning quality insights into completed cleanup work.

Cleanup Tasks

When a quality issue is identified, create a cleanup task assigned to the record owner:

• The task includes the specific issue, the affected record, and the suggested fix.
• Tasks appear in the rep's standard task queue alongside sales activities.
• Completing the task (by fixing the data) automatically resolves the quality issue and updates the score.

Cleanup Sprints

For larger data quality initiatives, create time-boxed cleanup sprints:

1. Define the scope (e.g., "All accounts with quality score below 50" or "All deals missing close dates").
2. Set the sprint duration (typically 1-2 weeks).
3. Distribute tasks across the team based on record ownership.
4. Track progress on the sprint dashboard showing completed vs. outstanding tasks.
5. Celebrate completion and measure the impact on overall quality scores.

Bulk Actions

For systemic issues affecting many records, use bulk actions:

• Bulk field updates: Update a field across all records matching a filter (e.g., set industry to "Technology" for all accounts with @tech-domain.com emails).
• Bulk merge: Review and accept multiple duplicate merge suggestions in a queue view.
• Bulk archive: Archive records that are irreparably low quality and polluting your data (e.g., test records, spam leads).

---

Best Practices

Follow these guidelines to build and maintain a culture of data quality across your sales organization.

• Establish quality standards early. Define required fields, naming conventions, and validation rules before your CRM has thousands of records. Retroactive cleanup is ten times harder than upfront prevention.
• Make quality visible. Display quality scores on record list views and individual record pages so reps see the impact of incomplete data entry in real time, not weeks later in a report.
• Tie quality to outcomes. Share data showing that deals with quality scores above 80 close at 2-3x the rate of deals scoring below 60. When reps see the correlation between data hygiene and commission, behavior changes.
• Address decay weekly. Spend 15-30 minutes each week reviewing and actioning quality suggestions. Small, frequent efforts prevent the backlog from becoming overwhelming.
• Use stage gates. Require minimum quality scores before a deal can advance to the next pipeline stage. A deal cannot move to "Proposal" unless it has a linked contact, a decision-maker identified, and a validated close date.
• Automate where possible. Use enrichment integrations to automatically populate firmographic data, validate email addresses, and update job titles. Manual data entry should be the exception, not the rule.
• Review imports carefully. Bulk data imports are the most common source of quality degradation. Always validate import files against your quality rules before importing, and quarantine records that fail validation.
• Celebrate quality leaders. Recognize reps who maintain high quality scores in team meetings. Positive reinforcement drives better habits than punitive measures.
• Audit quarterly. Even with automated monitoring, conduct a quarterly manual review of quality rules, thresholds, and staleness cadences. Business processes change, and your quality definitions should evolve with them.
• Start with high-impact records. Focus quality improvement efforts on active deals and engaged leads first. Cleaning up closed-lost opportunities from two years ago can wait until the records that drive current revenue are pristine.