Published Work
Free research on fintech, payments, and cloud infrastructure. Practical guides on AI adoption and productivity.
Research & Guides
Build Your First AI Agent With Google Gemini Gems
A step-by-step playbook for building a custom AI agent — no coding required. Complete walkthrough included.
Read → PaymentsPayment Orchestration Is Eating the Stack
How orchestration platforms are inserting themselves between merchants and PSPs — and restructuring the payments stack.
Read → AI AdoptionA Non-Technical Guide to Choosing Your First AI Tools
A practical framework for choosing your first AI tools — based on what you actually do, not what's trending.
Read → AI Adoption5 AI Adoption Mistakes That Waste Time and Money
The most common AI adoption mistakes professionals and teams make — and what to do instead.
Read →What a 48-Hour Market Snapshot Looks Like
Payment Orchestration Layer Economics for a Series B Fintech
8-page research memo · Delivered in 41 hours
Memo Structure
- Executive Summary — Key findings and recommendation
- Market Context — Current state of payment orchestration
- Competitive Landscape — 12 players mapped by positioning and traction
- Unit Economics Analysis — Interchange splits, take rates, margin structure
- Structural Risks — Regulatory, platform dependency, commoditization vectors
- Recommendation — Go/no-go assessment with supporting logic
- Sources & Methodology — 34 sources cited
Excerpt from Sample Memo
From the Executive Summary:
The payment orchestration layer is consolidating around three pricing models: per-transaction fees (2–5¢), monthly platform fees ($2K–$15K), and hybrid models that blend both. For a Series B fintech processing 800K monthly transactions, the orchestration layer represents a 3–7 basis point cost addition — offset by an estimated 12–18 basis point reduction in payment failure costs through intelligent routing and automated failover.
The core strategic question is not whether to adopt orchestration, but when the transaction volume justifies the integration cost. Based on the analysis below, the break-even point sits at approximately 400K monthly transactions for a multi-PSP setup, or 1.2M for merchants currently single-homed on Stripe or Adyen.
From the Unit Economics Analysis:
Spreedly’s published pricing of $0.03 per transaction at scale implies a $24K annual cost at 800K monthly volume — roughly 0.8% of a $3M annual payment processing spend. However, the total cost of orchestration extends beyond the platform fee. Integration engineering (estimated 120–200 developer hours for a standard REST API setup), ongoing PSP relationship management, and reconciliation complexity add approximately 40–60% to the direct platform cost in Year 1, declining to 15–20% in subsequent years as integrations stabilize.
What an AI Advisory Session Looks Like
AI-Augmented Content Workflow for a 4-Person Marketing Team
60-minute session + follow-up summary · Team of 4
Before the Session
- Writing blog posts end-to-end manually (~3 hours per post)
- Creating social media copy from scratch for each platform
- No AI tools in use — unclear which tools to try or trust
- Concerned about quality, brand voice consistency, and over-reliance
What the Session Covered
- Workflow audit — Mapped time spent per content task across the team
- Tool recommendations — Claude for long-form drafts, ChatGPT for social variants, Grammarly for brand voice consistency
- Process design — “Human-in-the-loop” workflow where AI generates first drafts, humans edit for voice and accuracy
- Prompt templates — 3 ready-to-use templates for blog outlines, social copy, and email subject lines
- Guardrails — What to review before publishing, and what AI handles well vs. poorly
Delivered in Follow-Up Summary
- Estimated time savings: 8–12 hours/week across the team
- Tool stack: 3 subscriptions recommended (total ~$80/month)
- Adoption plan: Week 1 trial → Week 2 expand → Week 3 evaluate
- Async support: One week of follow-up for troubleshooting and refinement