Case study · Cosmetics · Ingredient intelligence

RCMA Makeup: 163 analyst hours recovered in four weeks, with every number traceable to its source.

A four-week, no-fee proof-of-value pilot on live supplier documents — scored against acceptance criteria written before any pilot data was collected, and reviewed line by line by the client's own team.

Client

RCMA Makeup

The Research Council of Makeup Artists Inc.
RCMA Makeup · professional cosmetics

Workflow

Supplier document reconciliation
29 regulatory data categories

Pilot window

23 March – 16 April 2026
Four weekly sprints

Engagement

No-fee production-readiness pilot
Outcome: approved for production

The result
99.3%

n 3,480 human evaluations

Holistic extraction accuracy across 140 supplier documents.

163 h 20 m

n 140 documents · four weeks

Analyst time recovered over the pilot window.

2.4 min

145.9 s average · n 140

Processing time per document, ingestion to record.

$9,800

against $13.44 operating cost

Avoided processing cost over the pilot, estimated on the stated assumption — not cash recovered. Platform operating cost: $13.44; implementation and deployment investment separate.

The client and the problem

One trusted record, assembled by hand.

RCMA Makeup qualifies every raw material against the documents its suppliers provide. Ingredient intelligence is the work of turning those documents into one record the business can rely on.

Certificates of analysis, technical and safety data sheets, supplier regulatory statements — each carrying part of the picture, across 29 regulatory data categories per ingredient. Done by hand it is careful, repetitive reading: open the file, find the statement, retype the value, compare it against the previous revision and against every other document in the folder.

The hardest cost to see is what happens when documents disagree. A conflict stops being one analyst's task and becomes a small cross-functional investigation — R&D, regulatory, quality, and operations reading the same paragraph against each other, at times drawing in senior management. RCMA's baseline was set with that burden on the table.

Stated assumption

The savings baseline — 70 minutes of analyst work and $70 of loaded cost per document — was set by RCMA consensus, prompted by a discussion that surfaced the hidden cost of those cross-functional investigations, which often drew in senior management. It is published as an assumption, not a measurement; every time and cost figure on this page rests on it.

The method

Criteria before evidence.

The acceptance criteria were written into the Pilot Plan in March 2026, before a single pilot document was processed.

Targets and lower mandatory floors were anchored on 18 development-sample documents, so the thresholds could not be chosen to flatter the result. That left the pilot one question to answer: does performance hold at production volume, on documents the system has never seen?

The four weeks ran as four sprints. Documents were processed Monday to Thursday; analysis and refinement happened Friday to Sunday. Two RCMA users met a stable system every working morning — improvement happened on weekends, not in their laps.

Mon–Thu process Fri–Sun analyse & refine × 4 sprints
The pilot ran with the rigor of a Six Sigma validation process. Acceptance criteria were defined before development began, every result was traceable to its source, and everything was backed by evidence.
Agi Denes, Ph.D., MBA · R&D Director
The Research Council of Makeup Artists Inc.

SIPOC · Manual ingredient intelligence workflow · improved by Elara

Suppliers
Inputs
Process
Outputs
Customers
  • Ingredient suppliers / technical contacts
  • RCMA R&D team
  • RCMA regulatory / business stakeholders
  • Supplier ingredient documentation
  • Existing ingredient records
  • RCMA data-category schema
  • RCMA business rules / interpretation context
  • Excel-ready ingredient intelligence record
  • Source references & supporting statements
  • Gap / conflict flags
  • Supplier follow-up needs
  • Updated ingredient database
  • R&D
  • Regulatory / compliance
  • Product development
  • Operations / supplier-facing users
Receive & organize documents
Review & interpret source content
Extract, summarize & categorize data
Reconcile across documents and prior records
Identify gaps, conflicts & follow-up needs
Update record for human review and downstream use
Figure 1   The ingredient intelligence workflow as mapped before the pilot — suppliers, inputs, the six process steps, outputs, and the internal customers of the record. The acceptance criteria were written against this map.
How it ran

Elara flags. People decide.

Elara is an advisory system. It reads, extracts, reconciles, and assesses completeness; it does not make regulatory decisions. Human review precedes any formal use of its output, and every record it produces carries that statement on its face.

01

Ingest

Supplier documents arrive in batches and are classified by type before anything is read for content.

02

Extract

Each regulatory data category is captured with its source document and the exact supporting statement attached.

03

Reconcile

Compatible data is merged; conflicts across documents and against the prior record are surfaced with the reasoning shown.

04

Assess

Completeness is scored per category, and gaps, conflicts, and compliance-critical items are flagged for follow-up.

What one field looks like

Illustrative · synthetic data

Category Heavy metals — lead (Pb)
Value Not more than 10 ppm
Source Certificate of analysis, lot A‑2291 · p. 2
Supporting statement “Lead (Pb): not more than 10 ppm, USP <233>.”
Flag Conflicts with prior record (5 ppm) — human review required before use.
The acceptance record

Seven criteria, set in advance. Seven results.

Critical-to-quality criteria · Elara pilot, RCMA Makeup

All targets met

Criterion Mandatory Target Result Status
Holistic extraction accuracy Every data category assessed on every document — including the cases where the correct answer is that the source says nothing. ≥ 90% ≥ 95% 99.3% n 3,480 Met
Capture extraction accuracy Values that are stated in the document, captured correctly. ≥ 85% ≥ 90% 96.0% n 651 Met
Document classification Each document assigned the right type before its content is read. ≥ 90% ≥ 95% 95.0% n 120 Met
Reconciliation accuracy Matches and conflicts across documents and prior records resolved correctly, with the reasoning shown. ≥ 85% ≥ 90% 93.8% n 209 Met
Task evaluation accuracy The completeness judgement the system makes per task, checked against the reviewers. ≥ 85% ≥ 90% 98.6% n 355 Met
Cycle time per document Wall-clock processing time from ingestion to updated record. < 5 min avg 145.9 s n 140 Met
Data-integrity failures Any loss, corruption, or mis-assignment of a value during processing. None permitted None observed in 140 docs Met

a  Holistic accuracy counts the cases where the correct answer is that the source says nothing — the check that the system does not invent values.

b  Figures as recorded in the Elara Pilot Final Report, issued May 2026, covering the window 23 March – 16 April 2026.

The challenge

In Sprint 2, classification fell below its floor.

89.1%

Sprint 2 classification

100%

Confirmation run · n 20 × 2

Document classification accuracy dipped to 89.1% in the second sprint — under its mandatory floor.

The root cause was ambiguous handling of signator authority on supplier regulatory statements: the system was not reliably distinguishing whose voice a document represented, and mis-classified those documents as a result.

The fix was a designed series of experiments, not a patch. Candidate prompts, models, and parameter settings were screened systematically against the set of samples that had failed, and the best-performing combination was confirmed against that same set before returning to production volume.

We read the recovery in later sprints with a healthy reservation: those figures alone cannot prove the fix, because later document batches may simply have been easier. The screening against the failing set is the evidence; the sprint trajectory is only consistent with it.

Long term, the outcome is stabilized the way a process should be: a control phase in production — human-in-the-loop review and targeted spot-checks — holds the gain rather than a one-time fix.

Verification

Audited, not self-reported.

The accuracy figures are not model self-assessments. Every data point was scored by client reviewers and audited line by line, and nothing was counted until reviewer and auditor agreed. AI-assisted evaluations were themselves manually audited, and the full trail was retained.

The audit pass was performed by Smart-Suited Tech alongside RCMA's reviewers, so we do not describe it as independent. We describe it as documented, and the record is available on request.

Reviewer consensus
Audit trail retained
Source-linked fields
Human review required
In their words
The results spoke for themselves: 99.3% accuracy across 140 of our supplier documents.
Agi Denes, Ph.D., MBA · R&D Director
The Research Council of Makeup Artists Inc.
I would highly recommend them to any organization looking for an experienced, professional AI development team.
Audra Sanchez · Director of Operations
The Research Council of Makeup Artists Inc.
What happened next

Approved for production.

All mandatory floors cleared. All targets met. No data-integrity failures observed, and no critical defects open at close.

RCMA approved production deployment. Elara now runs in RCMA's own cloud tenant: RCMA controls the infrastructure and the spend, with US data residency configurable. Monitoring continues through the human review step and targeted spot-checks — the same review the pilot was built around.

Delivered by Smart-Suited Tech, then operating as Automa Services LLC.

The same pilot RCMA Makeup ran is available to you.

Four weeks, one real workflow, acceptance criteria agreed before we process a document. Share the workflow and we'll tell you honestly whether we can help.

Case study · Cosmetics · Ingredient intelligence

RCMA Makeup: 163 analyst hours recovered in four weeks, with every number traceable to its source.

A four-week, no-fee proof-of-value pilot on live supplier documents — scored against acceptance criteria written before any pilot data was collected, and reviewed line by line by the client's own team.

Client

RCMA Makeup

The Research Council of Makeup Artists Inc.
RCMA Makeup · professional cosmetics

Workflow

Supplier document reconciliation
29 regulatory data categories

Pilot window

23 March – 16 April 2026
Four weekly sprints

Engagement

No-fee production-readiness pilot
Outcome: approved for production

The result
99.3%

n 3,480 human evaluations

Holistic extraction accuracy across 140 supplier documents.

163 h 20 m

n 140 documents · four weeks

Analyst time recovered over the pilot window.

2.4 min

145.9 s average · n 140

Processing time per document, ingestion to record.

$9,800

against $13.44 operating cost

Avoided processing cost over the pilot, estimated on the stated assumption — not cash recovered. Platform operating cost: $13.44; implementation and deployment investment separate.

The client and the problem

One trusted record, assembled by hand.

RCMA Makeup qualifies every raw material against the documents its suppliers provide. Ingredient intelligence is the work of turning those documents into one record the business can rely on.

Certificates of analysis, technical and safety data sheets, supplier regulatory statements — each carrying part of the picture, across 29 regulatory data categories per ingredient. Done by hand it is careful, repetitive reading: open the file, find the statement, retype the value, compare it against the previous revision and against every other document in the folder.

The hardest cost to see is what happens when documents disagree. A conflict stops being one analyst's task and becomes a small cross-functional investigation — R&D, regulatory, quality, and operations reading the same paragraph against each other, at times drawing in senior management. RCMA's baseline was set with that burden on the table.

Stated assumption

The savings baseline — 70 minutes of analyst work and $70 of loaded cost per document — was set by RCMA consensus, prompted by a discussion that surfaced the hidden cost of those cross-functional investigations, which often drew in senior management. It is published as an assumption, not a measurement; every time and cost figure on this page rests on it.

The method

Criteria before evidence.

The acceptance criteria were written into the Pilot Plan in March 2026, before a single pilot document was processed.

Targets and lower mandatory floors were anchored on 18 development-sample documents, so the thresholds could not be chosen to flatter the result. That left the pilot one question to answer: does performance hold at production volume, on documents the system has never seen?

The four weeks ran as four sprints. Documents were processed Monday to Thursday; analysis and refinement happened Friday to Sunday. Two RCMA users met a stable system every working morning — improvement happened on weekends, not in their laps.

Mon–Thu process Fri–Sun analyse & refine × 4 sprints
The pilot ran with the rigor of a Six Sigma validation process. Acceptance criteria were defined before development began, every result was traceable to its source, and everything was backed by evidence.
Agi Denes, Ph.D., MBA · R&D Director
The Research Council of Makeup Artists Inc.

SIPOC · Manual ingredient intelligence workflow · improved by Elara

Suppliers
Inputs
Process
Outputs
Customers
  • Ingredient suppliers / technical contacts
  • RCMA R&D team
  • RCMA regulatory / business stakeholders
  • Supplier ingredient documentation
  • Existing ingredient records
  • RCMA data-category schema
  • RCMA business rules / interpretation context
  • Excel-ready ingredient intelligence record
  • Source references & supporting statements
  • Gap / conflict flags
  • Supplier follow-up needs
  • Updated ingredient database
  • R&D
  • Regulatory / compliance
  • Product development
  • Operations / supplier-facing users
Receive & organize documents
Review & interpret source content
Extract, summarize & categorize data
Reconcile across documents and prior records
Identify gaps, conflicts & follow-up needs
Update record for human review and downstream use
SIPOC diagram of the manual ingredient intelligence workflow. Suppliers: ingredient suppliers and technical contacts, the RCMA R and D team, and RCMA regulatory and business stakeholders. Inputs: supplier ingredient documentation, existing ingredient records, the RCMA data-category schema, and RCMA business rules. The process runs in six steps: receive and organize documents; review and interpret source content; extract, summarize and categorize data; reconcile across documents and prior records; identify gaps, conflicts and follow-up needs; update the record for human review and downstream use. Outputs: an Excel-ready ingredient intelligence record, source references and supporting statements, gap and conflict flags, supplier follow-up needs, and an updated ingredient database. Customers: R and D, regulatory and compliance, product development, and operations or supplier-facing users.
Figure 1   The ingredient intelligence workflow as mapped before the pilot — suppliers, inputs, the six process steps, outputs, and the internal customers of the record. The acceptance criteria were written against this map.
How it ran

Elara flags. People decide.

Elara is an advisory system. It reads, extracts, reconciles, and assesses completeness; it does not make regulatory decisions. Human review precedes any formal use of its output, and every record it produces carries that statement on its face.

01

Ingest

Supplier documents arrive in batches and are classified by type before anything is read for content.

02

Extract

Each regulatory data category is captured with its source document and the exact supporting statement attached.

03

Reconcile

Compatible data is merged; conflicts across documents and against the prior record are surfaced with the reasoning shown.

04

Assess

Completeness is scored per category, and gaps, conflicts, and compliance-critical items are flagged for follow-up.

What one field looks like

Illustrative · synthetic data

Category Heavy metals — lead (Pb)
Value Not more than 10 ppm
Source Certificate of analysis, lot A‑2291 · p. 2
Supporting statement “Lead (Pb): not more than 10 ppm, USP <233>.”
Flag Conflicts with prior record (5 ppm) — human review required before use.
The acceptance record

Seven criteria, set in advance. Seven results.

Critical-to-quality criteria · Elara pilot, RCMA Makeup

All targets met

Criterion Mandatory Target Result Status
{{ row.criterion }} {{ row.definition }} {{ row.floor }} {{ row.target }} {{ row.result }} {{ row.sample }} Met

a  Holistic accuracy counts the cases where the correct answer is that the source says nothing — the check that the system does not invent values.

b  Figures as recorded in the Elara Pilot Final Report, issued May 2026, covering the window 23 March – 16 April 2026.

The challenge

In Sprint 2, classification fell below its floor.

89.1%

Sprint 2 classification

100%

Confirmation run · n 20 × 2

Document classification accuracy dipped to 89.1% in the second sprint — under its mandatory floor.

The root cause was ambiguous handling of signator authority on supplier regulatory statements: the system was not reliably distinguishing whose voice a document represented, and mis-classified those documents as a result.

The fix was a designed series of experiments, not a patch. Candidate prompts, models, and parameter settings were screened systematically against the set of samples that had failed, and the best-performing combination was confirmed against that same set before returning to production volume.

We read the recovery in later sprints with a healthy reservation: those figures alone cannot prove the fix, because later document batches may simply have been easier. The screening against the failing set is the evidence; the sprint trajectory is only consistent with it.

Long term, the outcome is stabilized the way a process should be: a control phase in production — human-in-the-loop review and targeted spot-checks — holds the gain rather than a one-time fix.

Verification

Audited, not self-reported.

The accuracy figures are not model self-assessments. Every data point was scored by client reviewers and audited line by line, and nothing was counted until reviewer and auditor agreed. AI-assisted evaluations were themselves manually audited, and the full trail was retained.

The audit pass was performed by Smart-Suited Tech alongside RCMA's reviewers, so we do not describe it as independent. We describe it as documented, and the record is available on request.

Reviewer consensus Audit trail retained Source-linked fields Human review required
In their words
The results spoke for themselves: 99.3% accuracy across 140 of our supplier documents.
Agi Denes, Ph.D., MBA · R&D Director
The Research Council of Makeup Artists Inc.
I would highly recommend them to any organization looking for an experienced, professional AI development team.
Audra Sanchez · Director of Operations
The Research Council of Makeup Artists Inc.
Read the client stories in full
What happened next

Approved for production.

All mandatory floors cleared. All targets met. No data-integrity failures observed, and no critical defects open at close.

RCMA approved production deployment. Elara now runs in RCMA's own cloud tenant: RCMA controls the infrastructure and the spend, with US data residency configurable. Monitoring continues through the human review step and targeted spot-checks — the same review the pilot was built around.

Delivered by Smart-Suited Tech, then operating as Automa Services LLC.

Explore Elara How we handle your data

The same pilot RCMA Makeup ran is available to you.

Four weeks, one real workflow, acceptance criteria agreed before we process a document. Share the workflow and we'll tell you honestly whether we can help.