Vladimir Sadkov approaches a technology business as a sequence of tests, not a monument to an original idea. Start with a costly problem, prove that people want it solved, build the smallest credible product, and scale only after users return and pay. In a fast market, evidence matters more than enthusiasm.
Based in Dubai, Sadkov works across blockchain, AI, and fintech. Before turning to Web3, he built and sold two companies in EdTech and MedTech. His work has taken him across different products and markets, where what customers want, how a product reaches them, and the rules it must follow can change while it is still being built.
His method begins before product development, with a less flattering question: what are people doing without you?
Starting with the problem: how Vova Sadkov approaches new businesses
Founders naturally become attached to solutions. A new interface, model, protocol, or workflow feels tangible, so it attracts attention while the underlying problem receives less scrutiny. That order should be reversed.
You can tell a start-up idea has real commercial potential when you see people already spending resources to deal with the issue.
“I look at the problem instead of the idea itself. Maybe people are paying for a tool that barely does the job. Or maybe they are losing hours doing things by hand or bringing on extra headcount,” shares Sadkov. When you see those messy workarounds in the market, it tells you one very clear thing. That problem has a real financial cost attached to it.
This is also where market research becomes useful. The U.S. Small Business Administration’s guidance on market research asks founders to examine demand, market size, saturation, location, and the prices customers pay for alternatives. These are basic questions, but they force an idea into contact with an actual market.
The distinction matters because a person saying, “That sounds useful,” has paid nothing and risked nothing. Someone who spends every Friday reconciling data has revealed a real burden; agreeing to a pilot or paying a deposit sends a stronger signal.
As an entrepreneur, Vova Sadkov treats the idea as a hypothesis, the problem as the subject of the test, and the product as the instrument used to run it.
How to validate a start-up idea before building too much
Validation is not a survey that produces encouraging percentages. It is a chain of increasingly expensive commitments, with each step reducing a specific uncertainty.
Start with interviews, but ask about existing behaviour. “Would you use this?” invites politeness and imagination, while “How did you handle this last week?” produces facts. Useful follow-up questions cover frequency, cost, failed alternatives, decision makers, and the consequence of doing nothing.
Next, test the value proposition without building the complete product. Depending on the market, that might mean a landing page, a clickable prototype, a manual concierge service, or a narrowly scoped pilot. The format matters less than the response it measures.
The aim is not to prove the founder right, but to find the weak assumption while change is still cheap.
An MVP, or minimum viable product, supports that learning. “Minimum” does not mean careless; it means removing anything that does not test demand or deliver the core value. A fintech onboarding tool may need real security controls from day one, but not ten reporting views or a complex referral system.
Y Combinator’s explanation of product-market fit makes a related point: hiring, fundraising, and product polish do not prove fit. Strong demand appears in the behaviour of loyal, active, ideally paying users. The market begins to pull the product from the team.
Early signals can arrive quickly, but “quickly” is contextual. A consumer app can test repeat use within days, while an enterprise product may require procurement, integration, and security review. Founders should set a reasonable learning window before launch; without one, every weak result can be excused as “too early.”
Demand first, speed second, and quality according to risk
On an early team, demand, speed, and quality are often discussed as competing priorities. They are better understood as a sequence.
First, establish that a market wants the outcome. Then move fast enough to learn before assumptions become expensive, improving quality through repeated use and feedback. This order prevents months of work on a product that nobody needs.
Speed matters in business because it cuts down the time between making a decision and seeing the actual results. But moving fast is never a valid excuse for just shipping things randomly. A genuinely productive team operates differently. They zero in on a very specific question, put out a targeted test, and look closely at the data before they figure out the next move. Pushing out updates just to stay busy without trying to learn something specific does not help your bottom line. It just creates a lot of operational noise.
What cannot wait: security, privacy, and compliance
Quality has two layers: what a valid test requires and what protects users and the business. Visual polish may belong to the first layer, while security, privacy, transaction integrity, and legal compliance often belong to the second. Those protections cannot wait for product-market fit.
That distinction really matters when you start mixing AI, fintech, and blockchain. If you look at the NIST AI Risk Management Framework, they view risk management as an ongoing effort that has to happen during the design, development, deployment, and evaluation of any AI system. So if you are running an early stage company, you can absolutely cut back on features to get to market faster. But what you cannot do is ignore things like model risk, data handling, or potential user harm and hope to figure them out later when you have more resources.
Sadkov builds his products in the UAE and encounters VARA’s licensing requirements in practice. This brings regulatory questions into product planning: which activities the business will offer, where it will operate, and what permissions those activities require.
What early market demand actually looks like
Founders often look for one decisive metric, yet early demand rarely appears that neatly. It shows up as a pattern across acquisition, use, retention, and payment.
Four signals deserve close attention:
- Users arrive without being pushed. Referrals, direct searches, and unsolicited introductions suggest that the problem is worth discussing.
- Users return on their own. Repeat behaviour is stronger than registrations. It shows that the product has entered a real workflow or habit.
- Users notice failures. Complaints are uncomfortable, but indifference is worse. When a broken feature disrupts someone, the product has become useful.
- Users pay or make another costly commitment. Revenue, a signed pilot, shared operational data, or an internal rollout carries more weight than praise.
Feature requests can also be informative because they show that users imagine the product becoming more useful. Yet requests should not control the roadmap by vote. The team needs to find the repeated job behind them, since five different requests may reveal one deeper need.
These signals should be separated by customer segment. A blended retention number can hide one narrow group that loves the product while everyone else leaves; that group may be the real market.
Sadkov’s products have attracted substantial activity: one exceeded $100 million in trading volume within two weeks of launch, while another onboarded more than one million users, with over $900 million in connected wallet value. The next questions are about what continues—whether users keep coming back, how long they stay, and whether the product earns revenue from that use.
For Vladimir Sadkov, willingness to pay is the most honest signal because it forces a trade-off between this solution and another use of money. In a free pilot, the equivalent may be access to a team, data, integration work, or a scheduled rollout. Whatever the form, the commitment must have a cost.
How to scale a start-up without outrunning product-market fit
Scaling should follow repeatable demand, but too many companies reverse the order. They add marketing spend, roles, markets, and infrastructure to a model that has not settled. The result is a larger version of the original uncertainty, not a scalable business.
Investors including Animoca Brands, OKX Ventures, Lemniscap, Gate Ventures, and KuCoin Ventures have backed Sadkov’s ventures, which have raised more than $15 million in total. That capital gives the teams more options. Deciding which ones to pursue still requires evidence that demand can be sustained, rather than a funding announcement alone.
Premature hiring is one common symptom. A team recruits for the company described in its pitch deck, then spends time coordinating people whose work is not yet defined. Before fit, a compact team usually learns faster; specialists become valuable when the core motion is repeatable and the bottleneck is clear.
Loss of focus creates a similar problem. One strong use case can produce sharp messaging, faster onboarding, and comparable feedback, while five average use cases create five partial products. The team stays busy, but no customer group receives a complete solution.
Unit economics provide another guardrail. If you are running a company you need to know exactly what it costs to bring in a customer and provide your service. You should understand your gross margin and how long it takes to make your acquisition spend back. You also need to look at retention to see if your business model actually works in practice. Those early numbers are probably going to be rough estimates – that is normal. The real danger is choosing to ignore them entirely just because your top line revenue is going up. Running the business on imperfect data is always safer than ignoring the costs altogether.
You need to factor in regulatory requirements right from the start. They dictate entirely too much of your operation to leave for later. These rules impact where you can do business and who you can accept as a customer. They shape your product design and how you handle custody or data storage. They also directly affect your marketing strategy and overall operating costs. When you operate in a regulated market, treating compliance as an afterthought does not make your growth plan aggressive. It just makes it incomplete.
Before scaling, a founder should be able to answer five questions:
- Which customer segment retains best, and why?
- Which acquisition channel produces customers who keep using the product?
- Does the core transaction or subscription contribute positive value?
- Can the product deliver reliably as volume rises?
- Are the required legal, security, and risk controls in place for the next market?
If those answers remain vague, the next investment should usually buy learning rather than reach.
A practical operating loop for a fast-changing market
A durable technology business does not depend on predicting every change; it needs a short, honest feedback loop.
Begin with a costly customer problem and collect evidence from current behaviour. Build a narrow test, ask for commitment early, then measure return use and payment. Improve the product around the strongest segment, recheck security, regulation, and unit economics, and scale only what has become repeatable.
Working with this feedback loop completely shifts how you view a setback as a founder. When a launch underperforms, it is easy to just blame the marketing. But that weak response might point to something else entirely. You could be dealing with a minor product flaw, targeting the wrong customer segment, or putting out a confusing message. Sometimes it is just a matter of asking for too much setup upfront or launching at the wrong time. Your next move has to be driven by the actual evidence you see. If users are trying the product and not coming back, purchasing advertising is simply not going to fix the underlying problem.
Keeping that kind of discipline is what actually protects a business from getting swept up in industry hype. We hear a lot of noise about AI and blockchain right now. It is true that these tools can cut operational costs, automate certain decisions and manage complex transactions. They might even introduce entirely new models for ownership. But all of those technical capabilities are useless unless they directly improve a specific outcome for the customer. The technology you use is never the market itself. It is just one tool available to help you serve that market better.
Conclusion
Vladimir Sadkov’s path from idea to execution follows a simple operating rule: earn the right to invest more. A costly problem justifies an MVP; real use guides iteration, and retention with willingness to pay tells the team whether growth is justified. Scale comes after reliable economics, security, and compliance.
This approach is demanding because it leaves little room for comforting stories. It also gives founders a workable way through uncertainty. Vova Sadkov does not treat speed as haste or quality as decoration. Both serve the same goal: learning what the market needs, then building a business capable of delivering it repeatedly.
About Vladimir Sadkov
Vladimir Sadkov, also known as Vova Sadkov, is a technology entrepreneur based in Dubai. His background includes traditional IT, two EdTech and MedTech exits, and the development of Web3 products for an international market.
Leave a Comment