Sergii Gudkov once took a product job with an unusual compensation plan: no regular salary, only a percentage of monthly revenue. The arrangement turned every product opinion into a personal wager. At first, he behaved as if the title required him to generate the best ideas in the room. He listened lightly to colleagues and trusted his own creativity. Then a couple of months passed with little growth. The feedback arrived in the bluntest possible form - his income.
The revelation was not glamorous. The chief executive did not care whose idea won. Customers did not hand out points for originality. Gudkov's useful contribution was to collect ideas, test them and direct resources toward whatever worked. His own suggestions belonged in the same pile as everyone else's. He later distilled the episode into a line that explains much of his approach to work: “My job wasn't to be the idea person. It was to be the idea collector.”
That lesson traveled well. Gudkov's career has crossed software development, marketing, education technology, digital advertising, email infrastructure and payments. Publicly documented stops include Preply, Adwisely, Mailtrap and Railsware, along with Conversion42, the company he founded and led. He is now identified with CPOS in Ottawa, where supplied business information lists him as chief marketing officer. The job titles move between product and growth, but the underlying problem stays familiar: how does a team convert uncertain information into something customers will use and a business can sustain?
The Useful Humiliation of Watching
Another early lesson arrived in a customer interview. Gudkov watched a user struggle for ten minutes with something he had considered intuitive. He remembers nearly crying. Anyone who has built software knows this particular discomfort: the interface is obvious right up to the moment a real person meets it.
He took from the experience a distinction that sounds simple and changes the work. Quantitative research can reveal where people get stuck. It does not necessarily reveal why. The answer may sit in an interview, a support exchange or the replay of a frustrated click. Finding the right problem can be as valuable as solving it, because a team can otherwise spend a year polishing a misunderstanding.
Customer research sounds fancy until you realize it is mostly asking why five times in a row, like a persistent toddler.Sergii Gudkov on product practice
His advice to younger product managers starts lower to the ground than strategy decks suggest. Execute an approved idea well. Study how competitors handle small moments such as error messages and edge cases. Connect a feature to an outcome the business can recognize. Report trouble early, with options attached. This is less thrilling than declaring a vision, but it gives a new product manager the credibility and context required to shape one later.
Money Without the Costume
Gudkov is unusually direct about revenue. Customer obsession and commercial discipline, in his view, do not have to pull in opposite directions. A company that cannot make money eventually loses the ability to help anyone. A feature that has no plausible relationship to value is an expensive decoration, even when a team enjoys building it.
The point is not that every screen needs an immediate price tag. It is that product teams should be able to draw the line from work to value without hiding behind foggy language. At Adwisely, Gudkov's work included feature adoption, advertising returns and a pricing model tied to recurring revenue and customer value. At Preply, he helped launch a revenue-generating product while working on experience and conversion. Different markets, same discipline: a product becomes a business only when somebody's problem and somebody's willingness to pay meet.
The decision needs three live signals
Conceptual model, not measured performance. The useful decision is where evidence, economics and timing can support one another.
This also explains his skepticism toward easy wins. Gudkov calls low-hanging fruit a harvesting technique, not a strategy. A small task with little evidence may be easy without being ripe. He keeps a separate backlog and lets candidate fixes accumulate support from analytics, customer feedback and support tickets. Once an opportunity has earned confidence, a team can execute quickly. The speed comes at the end of validation, not in place of it.
When the Score Refuses to Decide
Frameworks are helpful when they make an argument visible. They become dangerous when a team uses the score to pretend the argument is over. Gudkov has described a product test in which displaying customers' first names increased a key metric while a smaller group of customers objected. The practice was legal. The reputation question remained. Leadership split between following the measured lift and respecting the discomfort.
This, to Gudkov, is what a real decision looks like: incomplete information, conflicting values and no formula that can absorb the responsibility. A product manager may not hold the final authority. The role is often to gather evidence, clarify tradeoffs, put credible options in front of the right people and make sure a decision actually happens. The manager still needs a view, because facilitation without a position can become drift.
The best decision is the one that gets made.Sergii Gudkov on choosing at the right speed
Speed matters here in two directions. Move too quickly and the team misses information that changes the choice. Move too slowly and the opportunity closes. Tools such as RICE, KANO and MoSCoW can organize discussion, but Gudkov notes that leadership intervention, communication gaps and shifting priorities often determine how a framework behaves in practice. The spreadsheet cannot repair trust between the people using it.
A Prototype Buys a Decision
In 2025, Gudkov carried that philosophy into a three-day experiment. For an Ottawa ProductCamp presentation, he built a prototype of a Mailtrap MCP server with generative AI. The objective was not production polish. It was a cost-effective proof that gave colleagues something concrete to judge. The prototype earned buy-in, and the engineering team later made the feature live.
The sequence contains a useful boundary for the current AI cycle. Generative tools can shorten the path from question to demonstration. They do not eliminate the difference between a demo and a dependable service. Gudkov's recent writing about small businesses makes the tradeoff concrete. For an owner unlikely to hire an expert, a rough AI-made campaign may compete with doing nothing. Yet a cheap tool that corrupts a database, sends nonsense to customers or harms a reputation can create a bill much larger than the freelancer it replaced.
Trust, then, becomes part of product economics. Gudkov makes a similar point when writing about employee incentives. If people choose cash now over promised upside, the company may be tempted to diagnose weak long-term thinking. A more revealing question is whether the organization has behaved in a way that makes waiting rational. Products, workplaces and AI systems all ask people to surrender a little control in exchange for future value. Credibility determines the exchange rate.
The Career Between Departments
Gudkov studied at Kyiv Polytechnic Institute from 2002 to 2008. His professional path since then resists a neat departmental box. The technical background matters, but so do marketing funnels, pricing, customer psychology and the politics of a launch. In one project, he coordinated more than 70 people across teams without being their manager. That is product leadership in its common form: responsibility distributed widely, authority held elsewhere.
His solution is communication at the level of daily practice. Map who needs to know what. Notice whether the chief executive cares about growth, engineering about technical debt and customer success about support volume. Frame the case for each audience while keeping the central claim consistent. Repeat back what you think you heard. Send the follow-up note. A grand skill such as stakeholder management becomes improvable only after it is broken into small behaviors.
The same plain-spoken style runs through his public writing, including a newsletter called Panda's notes. There are jokes about practicing magic to predict the future and blackmail as stakeholder management, but the humor usually sets up an unfashionable point: product work is systematic. Monday can be customer conversations. Tuesday can be competitor research. Wednesday can collect ideas. Thursday can turn them into options. Friday can test the smallest useful version.
That weekly rhythm is a better summary of Gudkov's career than any single title. Product and marketing meet in the demand for attention. Software and payments meet in the demand for reliability. AI and small business meet in the demand for trust. His contribution is to keep asking the prior question: what would have to be true for this feature, campaign or promise to deserve belief?
The framework can come next. First, the team has to look.