Discovery and concept
Interviews, market walk-throughs and assumption mapping that turn a hunch into a testable proposition and a sensible first scope.
The methodNew products, carefully made · London
Index of chapters
ventures@kavendra.org · Mon–Fri 09:00–18:00 UK
Chapter 00 · Prologue
Kavendra is a small London studio for new digital products. We take ideas from rumour to release: discovery, prototypes, mobile apps, websites and the launch that follows. We go first into uncertain ground, carry the right instruments, and come back with something real.
Propose an expeditionTest your idea’s weak point
Image: generated illustration of a lantern on a ridge. Not a photograph of our work.
Chapter I · The premise
The most expensive mistakes in a new product are made before anyone notices them: a guess about what people want that nobody tested, a technical assumption that only fails at scale, a launch plan that forgets how people actually find things. By the time the product is public, those mistakes are built into its foundations.
Kavendra works where the light is poor. We treat the early weeks of a product as an expedition: define the destination, list what could stop us getting there, and test the most dangerous of those things first with the lightest instrument that will give a trustworthy answer. Only then do we build properly, and when we build, we build with craft.
Chapter II · What we make
Interviews, market walk-throughs and assumption mapping that turn a hunch into a testable proposition and a sensible first scope.
The methodPaper, clickable or coded, chosen to answer one question quickly. Prototypes are tools for learning, not miniature finished products.
PrototypingiOS and Android products built with care for performance, accessibility and the small details that make an app feel trustworthy on first open.
Mobile workLaunch sites, web apps and early platforms that load fast, read clearly and can grow without being rebuilt.
Web workStore submissions, launch pages, release notes and steady care afterwards, answered during London working hours.
LaunchChapter III · Instrument one
Every new product rests on four kinds of belief. Set how sure you are about each, and the chamber shows which one is under most pressure, and the cheapest experiment that would relieve it.
How confident are you, honestly, in each of these? Drag each to where it sits today.
Riskiest assumption
Illustrative guidance, not a forecast.

Chapter IV · Craft
A rough prototype shown to the right person teaches more than a polished one shown to everyone. But rough does not mean careless. We make early work deliberately, labelling what is real and what is pretend, so the people testing it react to the idea rather than to broken details.
When the product moves to build, the craft rises: typography set for reading, motion used to explain rather than decorate, interfaces that remain usable with a screen reader, a keyboard or a shaking hand on a bus. We design to WCAG 2.2 AA and treat it as a floor.
On craftChapter V · The route
Leg one · typically two weeks
Conversations with the people the product is for, a look at what they use today, and a written map of assumptions ranked by danger.
Leg two · one to three weeks
The lightest prototypes that can test the most dangerous assumptions. Each test has a question, a method and a stated result before it runs.
Leg three · weeks to months
Building the first release in short increments. You hold a working build every fortnight and can stop, steer or continue at each.
Leg four · ongoing
Launch, then the quieter work of learning from real use, fixing what matters and deciding what comes next.
Chapter VI · Instrument two
Choosing the wrong kind of prototype wastes weeks. Answer three questions and the compass points to the lightest form that will give you a trustworthy answer.
Chapter VII · Launch
New products are judged within seconds of their store listing or landing page. We prepare App Store and Google Play submissions with accurate privacy disclosures and reviewer notes, write release notes people want to read, and build launch pages that say plainly what the product does and who it is for.
After launch, we stay close: watching for policy changes, fixing defects, updating dependencies and helping you interpret what early users are actually doing.
Launch and publishingChapter VIII · Principles
Chapter IX · Instrument three
If you cannot say what the product is in one sentence, nobody else will be able to either. Fill in the blanks and the forge assembles a positioning statement, then flags vague words.
Chapter IX½ · Companions
Founders with an idea and a little runway who would rather learn quickly than build blindly. Established organisations exploring a new service and wanting to test it without committing a large budget. Charities and public-facing bodies with a problem their current tools cannot solve. Teams with a prototype that worked in a demonstration and now needs to become something real people can rely on.
What they share is a willingness to be surprised. The early stages of a product often reveal that the most exciting feature matters least, or that the real customer is not the one first imagined. We bring that news carefully and early, because it is cheaper then than at any later point.
We are a poor fit if you need a fixed-price build from a finished specification with no room for discovery, or if the product depends on regulated financial, legal or medical judgement that should come from qualified advisers. We will tell you so in the first conversation and, where we can, suggest what kind of help to look for instead.
Chapter X · Field notes
Chapter XI · Signal
A sentence is enough to start. We reply during London working hours with the questions we would ask on day one, and an honest view on whether Kavendra is the right party to travel with.
Before you go further
This site keeps one essential record in your browser: this choice. If you accept optional storage, the Instruments remember your entries on this device. No analytics. No advertising. Nothing else.
What is stored