My name is Martin Koderisch. After a decade inside payments businesses, and a decade advising them, I now bring an outside-in view to executive teams and boards facing significant commercial decisions.
Engagements typically run four to six weeks at a fixed price, or longer depending on what you need. I start on the analysis instead of spending the first month learning your market on your time and budget. Nothing is adapted from a previous client.
To help frame the work, I offer five entry points, running from the strategic and financial questions through to operational ones: whether growth is real, what to fund, who the product is for, whether an AI advantage lasts, and how plans get sequenced. Each one can stand alone or be combined as required.
Built the strategic and valuation case for a pivot and replatforming decision, and presented it for board approval. Then designed the value creation plan and its roll-up to a C-suite dashboard, and the data and AI strategy for the replatforming.
B2B segmentation across 13 markets, connecting acquisition, retention and capability gaps to one investment view. Acquisition rates turned out to be broadly uniform, which meant retention was the real differentiator.
A private equity investor was looking at a pre-IPO round in an AI-native fraud platform, raised ahead of a NASDAQ listing. Led commercial due diligence on the unit economics, scalability and competitive durability, and modelled yield and margin over a seven-year hold. At Edgar, Dunn & Company.
When an engagement calls for a model or an AI prototype, I build it rather than commission it. A product and a set of working prototypes, built on my own time, and the reason I can say what these systems do and do not do well.
A strategic finance intelligence product for owner-managed businesses (strafi.co). It connects to the systems a business already runs on, accounting, payments, pipeline, reads them together, and tells the owner what needs attention while there is still time to act.
Python and Flask on PostgreSQL, with an LLM reasoning layer on the Anthropic API. Live integrations with FreeAgent, Xero, Stripe, HubSpot and Companies House, with more being connected. Built with AI coding tools, which is possible now in a way it was not three years ago.
Various tools in Python and Next.js, on the questions engagements keep raising: the economics of AI investment at scale, diagnostics for growth-stage businesses, and others.
These systems fail in a particular way. Given thin or inconsistent data, a model will still produce a fluent answer. Most of the work goes into deciding what the system is allowed to conclude, and showing the basis for every conclusion it reaches. That is the part I would want to see before trusting one of these in a business.
Where the thinking behind the engagements gets worked out in public.
Nearly a decade as a Principal at Edgar, Dunn & Company, the specialist payments consultancy, across more than 75 engagements for banks, fintechs, networks, private equity firms and enterprise merchants, and leading development of the growth and commercial excellence practice. Multiple pre-deal commercial due diligence mandates, and expert witness work on interchange and platform disputes.
Before that, a decade in senior operating roles at Mastercard, Citi and the fintech Luup. Co-author of Price Management in Financial Services (Springer).
I have worked on AI projects for most of the past decade, though until Strafi these were machine learning rather than LLM.
The practical effect is that I do not need a discovery phase to understand a payments business, or a platform whose economics are driven by payment flows. That is what makes short, fixed-price engagements possible.