Case studies

Results that are measured. Not implied.

Most agency case studies show pretty videos. The buyer's question: did this actually work? never gets answered. Lumaris's case studies start with the number.

Book a 15-min call
Active retainer client

Code with Antonio

Antonio Erdeljac · 420K YouTube subscribers

99%

Viewer retention through intro sections

Platform norm: 60–70%

13.4%

CTR on sponsored segments

Industry benchmark: 2–7%

50×

Social impressions growth

First 90 days of engagement

5.0/5.0

Clutch review rating

Verified

The problem

Antonio runs one of the largest full-stack development YouTube channels. 420K subscribers. Premium sponsorship inventory. Audience consists of professional developers building real applications.

The problem most agencies introduce on a channel like this isn't quality, it's accuracy. A demo that uses a deprecated React pattern or a TypeScript signature that wouldn't compile creates a comments-section disaster the host has to clean up. The hidden cost: the host becomes the technical editor.

What Lumaris produced

We produce Antonio's sponsored demo segments end-to-end. The brief is technical - APIs, CLIs, database integrations, full-stack feature builds. The deliverable includes the demo video, custom thumbnails, short-form social cuts, and the technical validation receipt confirming what was checked.

  • Sponsored demo segments (mid-roll, full integration with episode topic)
  • Custom thumbnails optimized for Antonio's channel design system
  • Short-form cuts for cross-platform distribution
  • Dev Validation Receipt per video (delivered alongside .mp4)

Measurable result

99% viewer retention through intro sections, meaning Antonio's audience doesn't drop off during the sponsorship segment, which is what every demo video's editor is actually selling. The average sponsorship segment retention rate on YouTube hovers around 60–70% on educational content. 99% is two standard deviations above the platform norm.

13.4% click-through rate on sponsored segments. The benchmark for educational technical content is 2–7%. Lumaris-produced segments hit 13.4% consistently. That's 2–7× the industry baseline.

50× increase in social impressions across short-form cuts in the first 90 days of the engagement. Antonio's content has compounding distribution because the short cuts continue to surface across platforms long after the primary upload.

Polaris Intro Short
Active retainer client 0:59

Polaris Intro Short

Short-form / Social

Polaris Intro
Active retainer client 4:54

Polaris Intro

Tutorial

Resonance Intro
Active retainer client 5:00

Resonance Intro

Tutorial

Active retainer since 2024. Production cadence: continuous sponsored content production aligned to Antonio's monthly release schedule.

They specifically tailor their services to my needs, my brand, and my audience.

Antonio Erdeljac CEO, Code with Antonio · Verified 5.0/5.0 Clutch review Verified 5.0/5.0 Clutch review
Former retainer client

Thu Vu

Content Creator · Educational YouTube · Netherlands

40%

Reduction in editing revision time

vs. prior agency relationship

100%

On-time delivery rate

Across all videos in engagement

5.0/5.0

Clutch review rating

Verified

The problem

Thu Vu produces long-form educational technical content for a YouTube channel focused on data and AI engineering. The editorial complexity isn't surface-level (b-roll, sound design), it's structural. A 30-minute educational video has 30 distinct concept introductions, each requiring visual and pacing decisions that can lose the audience if executed poorly. The prior agency relationship was technically capable but operationally slow. Revisions consumed time better spent on next-content planning.

What Lumaris produced

Long-form technical educational videos. Concept visualization. Pacing for sustained 30+ minute viewer attention. Multi-format export for cross-platform distribution.

  • Long-form educational technical videos (30+ minutes each)
  • Concept-visualization animation
  • Multi-platform format exports
  • Direct collaboration with Thu Vu's editorial workflow

Measurable result

40% reduction in editing revision time. The first draft arrived structurally aligned with Thu Vu's editorial preferences, eliminating the 2–3 revision cycles required to land structure with the previous editor.

100% on-time delivery rate across the engagement. This sounds like a baseline expectation. In B2B video production, it isn't. Late delivery is industry-standard. Hitting 100% over a multi-video engagement is operational discipline, not luck.

The team is exceptional at understanding my content and producing high-quality edits to my videos.

Thu Vu Content Creator · Verified 5.0/5.0 Clutch review Verified 5.0/5.0 Clutch review

Engagement period: January 2025 – February 2026. Engagement complete. Thu Vu is no longer an active retainer client.

Calibration edit example

What a Calibration actually produces

CodeScene approached Lumaris with footage. Before any retainer conversation, our editors produced five distinct executions in CodeScene's brand language. Five styles. Same source material. Five different editorial philosophies.

For paid customer engagements, the Calibration produces three executions. For CodeScene, all five are presented here because the visual range across styles demonstrates how editorially distinct each cut can be while remaining on-brand.

Below are all five CodeScene Calibration variants. All produced in CodeScene's brand language. Each built with a different level of seriousness, dynamics, speed, tempo, and feel.

CodeScene Calibration Edit - Style 01
CodeScene Calibration Edit - Style 01
CodeScene Calibration Edit - Style 02
CodeScene Calibration Edit - Style 02
CodeScene Calibration Edit - Style 03
CodeScene Calibration Edit - Style 03
CodeScene Calibration Edit - Style 04
CodeScene Calibration Edit - Style 04
CodeScene Calibration Edit - Style 05
CodeScene Calibration Edit - Style 05

All five videos remain CodeScene's. The Calibration mechanic is non-extractive, the client keeps the work regardless of what happens next. If they continue to a retainer, the €2,500 credits toward Month 1. If they don't, they keep the videos.

This is the entry point to every Lumaris engagement. Before a retainer is signed, before a production schedule is set, the editorial style is locked.

Start your Calibration →
A case study in hidden costs

14 engineering review rounds eliminated across 6 videos

hidden-review-cost.calc
14 review rounds × 2+ hours per round (minimum)= 28+ hours of direct engineering/DevRel/marketing timePlus: context-switching cost14 rounds × ~30 min reorientation per workflow re-entry= 7+ hours of productivity loss before/after each reviewPlus: concept-explanation overheadEach round requires guiding the editor through corrections~1 hour average to explain WHY a correction is needed= 14+ hours of editorial education time──────────────────────────────────────────Total internal time absorbed:  49+ hours minimumAt €100/hour fully-loaded:      €4,900+At €150/hour fully-loaded:      €7,350+Same 6-video scope produced by Lumaris:0 engineering review rounds0 hours of engineering time€0 hidden cost──────────────────────────────────────────

Anonymized at client request. The pattern is industry-typical.

A developer tools company engaged us to produce six videos. The agency they had used previously required 14 cumulative engineering review rounds across that scope, between two and three rounds per video.

Each review round consumed at least two hours of internal time. A senior engineer reading the technical claims. A DevRel team member flagging terminology drift. A marketing manager translating between editorial language and engineering correctness. None of that time appears on the video invoice. All of it appears in engineering velocity loss.

The Dev Validation Receipt mechanism eliminates the need for client-side engineering review. The validation happens internally before the draft leaves our studio. The client receives the receipt naming what was checked and what was corrected. Engineering team never enters the loop unless the client wants them to.

The 6-video scope completed without a single engineering review round on the client side. Engineering team back on product. Marketing team back on distribution. Content shipped on schedule.

The cost isn't on the video invoice. It's in the engineering hours your team is already paying. Lumaris's pricing replaces that hidden cost with a visible one, at a lower total.

lumaris_dev-validation-receipt.log
 ═════════════════════════════════════════════
LUMARIS STUDIO · DEV VALIDATION RECEIPT
═════════════════════════════════════════════

Client:      [Client redacted]
Project:     Multi-format content system / 14-round validation
Delivered:   2024-08
Editor:      Lumaris Studio
Validator:   Lumaris QA

─────────────────────────────────────────────
WHAT WE CHECKED
─────────────────────────────────────────────
 Hook strength vs. audience intent (per video)
 Audio consistency across batch
 Thumbnail–title alignment
 Retention curve mapped to script decisions
 Card + end-screen placement
 Description SEO structure
 B-roll density and pacing alignment

─────────────────────────────────────────────
FLAGS RAISED
─────────────────────────────────────────────
[Round 03]
Issue:  Hook delayed 11 seconds - misaligned with search intent
File:   [filename redacted]
Action: Re-edit flagged, script revised, re-delivered in 48h

[Round 07]
Issue:  Audio inconsistency across 4-video batch - volume floor mismatch
File:   [batch redacted]
Action: Normalised across full batch, re-delivered

[Round 11]
Issue:  Thumbnail text illegible at mobile size across 3 formats
File:   [thumbnails redacted]
Action: Escalated to client design team with specific recommendations

─────────────────────────────────────────────
CORRECTIONS APPLIED
─────────────────────────────────────────────
Total flags raised:         31
Total corrected:            28
Total escalated to client:  3

─────────────────────────────────────────────
Validated and approved for delivery.
───────────────────────────────────────────── 
Here is what to expect

Demonstration work

Lumaris produces demonstration videos for brands we want to work with. The technical accuracy is the same standard as paid engagements. The editorial discipline is the same. These brands did not commission this work, we produced it independently.

Why this matters: a demonstration video produced under no commercial pressure is the cleanest possible signal of technical capability. There's no editorial compromise to a brief, no client-side feedback loop, no scope constraint. What's shown is what we believe the work should be.

GitButler Demo
Demonstration work

GitButler Demo

XBOW Demo
Demonstration work

XBOW Demo

Kestra Demo
Demonstration work

Kestra Demo

Chromatic Demo
Demonstration work

Chromatic Demo

ClickHouse Demo
Demonstration work

ClickHouse Demo

Solo.IO Demo
Demonstration work

Solo.IO Demo

These are demonstrations. The work we produce when no client is in the room. To produce something like this for your product specifically, the entry point is the Calibration.

Start with a Calibration →
Your case study

The case study above could be yours, six months from now.

Antonio's 99% intro retention number didn't exist when we started working with him. The 40% revision-time reduction Thu Vu measured didn't exist either. Both numbers came from removing the translation layer between editor and product.

The first conversation is a 15-minute fit call. We come prepared with observations about your current content. You leave with a specific recommendation on whether more video production is worth the investment right now, or whether the timing is wrong.

If timing is right, the entry point is the Calibration. €2,500. Three variants. Fully credited toward Month 1 if you continue.

 + currently producing for 8 active retainer clients + 4 calibration slots open this month + founder personally reviews every audit submission