← JournalMemo Nº 012026.10.2010 MIN READ

Build Without Asking: Agile, Lean and Bootstrapping from Cord

In class we covered Agile, Lean Startup and bootstrapping as three separate topics. Building Cord taught me they’re the same conversation seen from three angles: how you work, how you know you’re on the right track, and how you pay for the track. This memo is what has worked for me, what hasn’t, and what I’d tell anyone about to start something.

Three questions, not three camps

The first thing to get is that they don’t compete. A company can, and almost should, use all three at once, because each one answers a different question.

AgileHow do I organize the work?Short cycles, value bit by bit.
LeanHow do I prove my idea works?Experiments before spending.
BootstrappingHow do I fund my growth?Own revenue before outside capital.
Fig. 1Three approaches, three different questions

Mix up the questions and you make bad calls. I’ve seen teams believe that “being agile” means not planning, or that “being lean” means shipping something broken. Neither is true.

Agile: what actually matters in the Manifesto

The Agile Manifesto was signed by 17 developers in 2001. It isn’t a tool; it’s a set of values. It favors individuals over processes, working software over comprehensive documentation, customer collaboration over contract negotiation, and responding to change over following a plan.

The key word is over. It doesn’t say processes or documentation don’t matter; it says which one wins when they collide. At Cord documentation matters a lot, because several computers and several working sessions touch the same repository, but never above the product working. Every push to production has to pass security checks; if it doesn’t, it doesn’t ship. That’s “working software” turned into a rule.

Scrum: the roles people mix up

Scrum splits work into sprints of one to four weeks, and every sprint ends with something that works. It has three roles: the Product Owner decides what gets built and in what order; the Scrum Master removes obstacles so the team can work; and the team builds the increment.

Its events form a loop:

Planningwhat gets doneDaily15 min, sync upReviewshow what was builtRetroimprove how we workincrement
Fig. 2A Scrum sprint, from planning to retrospective

The classic trap: the Review shows the product; the Retrospective improves the process. And the artifacts: the Product Backlog is everything pending, prioritized; the Sprint Backlog is what goes into this sprint; the Increment is what already works.

Why I run Cord more on Kanban than on Scrum

Scrum shines when you have a new product and a team with goals per cycle. Kanban shines when the work is a continuous flow: support, improvements, things that show up every day. Cord has a bit of both, but its day-to-day looks more like a flow. Every change that’s ready gets merged and deployed; it doesn’t wait for the end of a sprint.

To doOpen a new countryEmail templates
In progressRecurring paymentsKYC improvementsWIP ≤ 3: nothing new starts until one closes
DoneClosing linkCFDI 4.0Verifactu
Fig. 3An example of the flow: few things in progress, everything done actually done

The Kanban metric that has helped me most is the WIP limit (work in progress). Starting lots of things feels productive and is exactly the opposite. My limit is almost literal: I work with several development sessions in parallel and never more than about five at once. And a rule I copy from my own repository: what’s finished gets merged, what isn’t ready gets said out loud. No branches left hanging “for later”.

Lean Startup: Cord’s MVP was another product

Lean Startup, Eric Ries’s 2011 book, proposes validating ideas with experiments before overspending. Its loop is build → measure → learn, repeated until the idea holds up or the evidence tells you to change course.

Buildthe minimum to testMeasurewhat users doLearnpersevere or pivotvalidated learning
Fig. 4The Lean Startup loop: the goal is to learn fast, not to build a lot

Cord’s MVP wasn’t called Cord. It was a Flouvia Shopify app for B2B quotes: a minimal version that already solved something real (quoting with negotiated prices) inside stores that already existed. What we learned there was that the pain wasn’t Shopify’s, and it wasn’t only B2B: any business that closes sales with proposals suffers the same thing. Send a PDF, wait for the signature, chase the transfer, invoice by hand, reconcile in Excel.

In June 2026 Cord started as a standalone product. That was a pivot: not a whim, but a change of direction backed by what we saw. We went from an app inside Shopify to a horizontal platform, from proposal to payment, for any business.

The other piece of validated learning that marked me was in Latin America. Cord Payments runs on Stripe Connect, which doesn’t cover every country we wanted to offer the same way. Instead of promising what we couldn’t sustain, in Brazil, Colombia, Argentina, Chile and Peru online payment runs through the business’s own Mercado Pago account. Deciding that was listening to the evidence, not to ambition.

What we didn’t have to build

A fintech doesn’t need to stand up its entire infrastructure to validate. Class sums it up in four pieces:

KYC means Know Your Customer: verifying who the customer is before moving their money. Building it from scratch would have taken months. With the right API, it’s one screen inside Cord.

Bootstrapping: growing on what you generate

Bootstrapping is funding growth with your own resources and reinvested revenue instead of outside capital. Venture capital is the opposite: investor money in exchange for part of the company.

CONTROLBootstrappingSPEEDVenture capitalit depends on how much money you need, how fast, and how much control you want to keep
Fig. 5Bootstrapping versus venture capital: there's no universal answer

Cord grows on what it generates: it’s at $5k MRR today and that revenue funds the next step. That forces a discipline no class teaches quite the same way. Every cost has to make sense against revenue, which is why almost everything in the infrastructure is pay-as-you-go. The business model is designed for it: a freemium subscription plus a share of the volume processed through Cord Payments. When the customer does well, Cord does well.

I’m not saying venture capital is bad. If your business needs a lot of money before it charges its first dollar, you probably need it. But if you can charge early, keeping control is worth a lot.

What I take with me

  1. Agile, Lean and bootstrapping answer different questions. Use them together.
  2. If you’re a small team, Kanban with a WIP limit beats textbook Scrum.
  3. Your first MVP can be a different product. Listen to what people use, not what they say.
  4. Don’t build what an API already solves. Build what makes you different.
  5. Charge early. Your customers’ money is the cheapest money you’ll ever get.