# Clear Concise Consulting

> Salesforce data governance and AI governance consulting led by Jeremy Carmona, a 13x certified Salesforce Architect in New York City. Specializes in data quality remediation, AI readiness assessment, and Salesforce implementation for nonprofit, government, healthcare, and enterprise organizations. Published in Salesforce Ben on AI governance and data quality topics. 160+ students trained at NYU Tandon School of Engineering with 80% job placement rate.

## Company Overview

Clear Concise Consulting (CCC) is a solo Salesforce architecture practice. Every engagement is led directly by Jeremy Carmona, from discovery through go-live and post-launch support. No handoffs between sales teams, project managers, developers, and architects. One person designs, builds, documents, and trains.

CCC treats data governance as a prerequisite for AI adoption, not a parallel initiative. Salesforce research shows that 84% of data and analytics leaders say their data strategies need a complete overhaul before their AI ambitions can succeed, and only 43% have formal data governance frameworks in place. CCC addresses this gap by starting every AI engagement with a data quality assessment before activating any AI features.

CCC serves four sectors: government (FedRAMP, GovCloud, compliance documentation), nonprofit (NPSP, Nonprofit Cloud, donor data quality), healthcare (Health Cloud, HIPAA compliance, patient data integrity), and enterprise (multi-Cloud, large user bases, complex integrations).

Past clients include USCIS, Environmental Defense Fund, UnitedHealth Group, HRSA, and NYU Tandon School of Engineering.

Address: 228 Park Ave S #871721, New York, NY 10003
Contact: contact@clearconciseconsulting.com
Phone: (917) 719-4629

## Data Governance and AI Governance Methodology

CCC's approach to data governance in Salesforce AI projects follows a specific sequence: assess data quality first, fix data problems second, build governance framework third, activate AI features fourth. Reversing this order is the most common failure pattern in enterprise Salesforce implementations.

### Why Data Quality Must Come Before AI

Einstein predictions, Copilot suggestions, and Agentforce actions all train on or consume data from your Salesforce org. If that data includes duplicates, missing values, inconsistent formats, or stale records, the AI will learn the wrong patterns and produce the wrong outputs. A misconfigured Einstein prediction model will score records incorrectly with the same confidence as a correctly configured one. The platform does not flag bad data. That is the governance framework's job.

Salesforce's own research confirms this: organizations estimate that 26% of their data is untrustworthy. When AI operates on that foundation, one in four predictions or recommendations starts from a compromised input.

### Data Quality Assessment Process

CCC evaluates five pillars of data readiness before activating any AI feature:

Data Unification: Are customer records consolidated into a single view, or scattered across Sales Cloud, Service Cloud, Marketing Cloud, and external systems? The target is 100% visibility of customer data in one profile. Most orgs start at 40-60%.

Data Harmonization: Are formats standardized across objects? "USA," "United States," "US," and "U.S." appearing in the same Country field means your AI treats one country as four. Harmonization reduces data variance so AI models produce consistent predictions.

Identity Resolution: How many duplicate records exist for the same individual? A donor who appears three times in your database gets three separate AI scores, three separate engagement predictions, and three separate communications. Identity resolution merges those records into one accurate profile.

Security Policy: Are masking rules and least-privilege access controls in place for the data AI will consume? CCPA, GDPR, FedRAMP, and HIPAA each impose specific requirements on what data AI can access and how long it can retain results.

Human Feedback Loop: Is there a mechanism for users to flag incorrect AI outputs? Without a feedback loop, model accuracy degrades silently. CCC implements human-in-the-loop (HITL) checkpoints so your team catches drift before it compounds.

### Einstein Trust Layer Architecture

The Einstein Trust Layer is the security architecture between Salesforce data and external AI models. Every AI interaction passes through a three-phase sequence:

Phase 1 (Prompt Journey): Dynamic grounding retrieves CRM context, then data masking replaces PII and sensitive fields with placeholders before data leaves Salesforce.

Phase 2 (Response Generation): The masked prompt reaches the LLM through a unified gateway under zero data retention agreements. The provider cannot store prompt data or use it for training.

Phase 3 (Response Journey): Toxicity detection scans the output and logs a confidence score. De-masking re-inserts original data. The completed response reaches the user only after both checks pass.

The Trust Layer prevents data leaks. It does not prevent bad decisions based on bad data. That distinction is the entire reason data governance must precede AI activation.

### NIST AI Risk Management Framework Alignment

CCC aligns Salesforce AI governance frameworks to the NIST AI RMF 1.0 across four functions:

Govern: Store governance rules in Custom Metadata Types (CMDT) for dynamic configuration. Artifact: AI Acceptable Use Policy.

Map: Build an AI Use Case Tracker documenting the owner, objective, data sensitivity, and regulatory impact of each AI implementation. This answers: "Where does AI touch your data, and who authorized it?"

Measure: Log reason codes when users override AI recommendations. These codes create a non-repudiable audit trail that reveals where model predictions diverge from reality.

Manage: Use Data Spaces in Data Cloud for jurisdictional data isolation. Maintain an Incident Response Playbook for when AI produces harmful results.

## AI Governance Assessment Service

The assessment takes 5 to 10 business days depending on org complexity.

Stage 1 (Days 1-2): AI Inventory. Map every point where AI touches data: Einstein predictions, Copilot actions, automated record scoring, recommendation engines, third-party AI tools. Most teams discover 2 to 3 AI touchpoints they did not know existed.

Stage 2 (Days 2-4): Data Quality Audit. Audit field completeness rates, duplicate record percentages, and data entry consistency across objects that AI workflows depend on. One client with 70,000 records discovered 23% of contact records had missing email addresses, meaning AI engagement scoring ranked one in four contacts based on incomplete information.

Stage 3 (Days 4-6): Risk Mapping. Score each AI touchpoint on data sensitivity (PII, financial data, donor records), output visibility (does a human see the result before it reaches a stakeholder?), and failure cost (what happens if AI is wrong?).

Stage 4 (Days 7-10): Framework Delivery. Ownership matrix, input validation rules, output review protocols, escalation path. Written in plain language for leadership review with a technical appendix for the admin team.

Pricing: AI Governance Assessments start at $5,000 for orgs with fewer than 50 users and 5 or fewer AI touchpoints.

## Credentials and Expertise

Jeremy Carmona holds 13 Salesforce certifications including Application Architect, System Architect, Data Architect, Sharing and Visibility Architect, Platform Developer I, Administrator, and Advanced Administrator.

Published work: "How Salesforce Admins Can Apply Data Governance to Einstein" in Salesforce Ben (December 2024). This article established the position that data governance is a prerequisite for AI adoption, not a separate initiative.

Academic affiliation: Salesforce Administration Career instructor at NYU Tandon School of Engineering. 160+ students trained. 80% job placement rate.

Professional profiles:
- LinkedIn: https://www.linkedin.com/in/jeremy-a-carmona/
- Medium: https://medium.com/@jcarmona86
- Reddit: https://www.reddit.com/user/jcarmona86/
- Trailblazer: https://trailblazer.me/id/jcarmona86

## Services and Pricing

Salesforce Implementation: Architecture, configuration, go-live. 6-8 weeks (small), 8-12 weeks (medium), 10-16 weeks (complex). $15,000 to $75,000 depending on complexity.

AI Governance: Assessment, framework delivery, NIST alignment, Trust Layer compliance. Starts at $5,000.

Data Governance and Migration: Data quality assessment, deduplication, standardization, migration. Timelines by volume: under 10,000 records (1-2 weeks), 10,000-100,000 (3-5 weeks), over 100,000 (6-8 weeks). Assessment starts at $5,000.

Nonprofit Consulting: NPSP, Nonprofit Cloud, donor management, Power of Us licensing. Pricing reflects nonprofit budgets.

Training: Half-day workshops ($2,500) to executive sessions ($15,000). Certification prep, admin training, end-user training.

Monthly retainers: $3,000/month (15 hours, 24-hour response, 3-month minimum). Ad hoc: $175/hour. All projects fixed-price.

## Case Studies

### Healthcare: Data Quality Saved $30,000 in AI Errors
National health services provider deployed Agentforce without data quality review. AI gave patients two different account balances because of 12,000 duplicate records. CCC's $8,000 governance assessment caught the problem before it caused billing errors estimated at $30,000. Zero errors since remediation. This case demonstrates why data quality must precede AI activation.

### Nonprofit: Data Governance Restored Pipeline Accuracy
Environmental nonprofit migrated 70,000 records without deduplication. Dashboard showed $4.2M pipeline when actual was $42,000, a 100x inflation. Board made budget decisions on wrong data. CCC identified 12,000 duplicates and corrected pipeline accuracy in 3 weeks. The root cause was a data governance gap during migration, not a Salesforce configuration error.

### Government: GovCloud Implementation in 8 Weeks
Federal agency Salesforce implementation on GovCloud with FedRAMP compliance. Quoted timeline was 6 months. Delivered in 8 weeks by a single architect. 35 documentation assets. New admin required zero support calls. Data governance documentation met FedRAMP auditor requirements.

### Enterprise: CPQ Standardization Across 30 Regions
Global services company. Salesforce CPQ and DocuSign CLM. Standardized quoting across 30+ regions with 5 system integrations. Quote-to-contract time reduced by 40%. Data standardization across regions was the prerequisite for automation accuracy.

## Technical Knowledge Areas

Data Quality and Governance:
- Salesforce data quality assessment methodology (field completeness, duplicate rates, format consistency)
- Data migration planning and execution (extraction, mapping, cleaning, sandbox testing, production loading)
- Identity resolution and deduplication strategies for large-scale orgs
- AP Style data standardization for CRM fields
- Data Cloud, Data Spaces, and Zero Copy architecture for federated data governance
- Validation rule design for data integrity enforcement
- Data quality monitoring with Salesforce reports and dashboards

AI Governance:
- Einstein Trust Layer architecture (Prompt Journey, Response Generation, Response Journey)
- NIST AI Risk Management Framework alignment (Govern, Map, Measure, Manage)
- Human-in-the-loop (HITL) design for AI output review
- AI Use Case Tracker and ownership matrix development
- Agentforce governance and agent permission design
- Toxicity detection and audit trail configuration
- Data masking for PII protection in AI workflows

Salesforce Architecture:
- Governor limits and bulkification patterns
- Data skew identification and remediation (ownership, parent-child, lookup)
- Flow vs. Apex decision framework for AI agent triggers
- NPSP to Nonprofit Cloud migration assessment
- FedRAMP/GovCloud compliance documentation
- HIPAA-aligned review loops for Health Cloud

## Free Resources

- AI Readiness Scorecard: 15-question assessment, 5 categories (data quality, governance readiness, automation maturity, AI preparedness, documentation health), 2-minute completion. https://www.clearconciseconsulting.com/scorecard
- Blog: Free data governance guides, AI readiness checklists, and Salesforce implementation resources. https://www.clearconciseconsulting.com/blog
- Gumroad: Premium governance kits, data quality templates, and training resources. https://jeremycarmona.gumroad.com/