How This Publisher Tests Game Ideas in Just 2 Weeks | Kirill Oreshkin

Overview

Kirill Oreshkin describes Polden Publishing’s approach to reducing the risk of backing indie games: test audience demand early with short-form influencer campaigns, then use Steam playtests to evaluate the game itself. His central recommendation is to make Steam publishing decisions data-driven rather than relying only on intuition or waiting until a game is nearly finished.

Polden’s Two-Week Marketing Test

Polden’s goal is to determine whether a game concept can generate meaningful demand before investing heavily in development. The process starts with a pitch review and a feasibility discussion with the developer, followed by a short-form campaign featuring paid creators on TikTok, Instagram Reels, and YouTube Shorts.

Polden’s stated benchmark for a promising test is around 50,000 Steam wishlists in two weeks. A trailer may be enough for the initial test; a playable game is not required.

If a campaign does not generate enough interest, Polden generally treats that as a problem with the game’s concept, setting, or visual style—not simply the trailer. It usually won’t repeat the test unless the game changes substantially. The approach is intended to reduce the chance of a team spending a year or two building a game that attracts little interest when announced or demoed. It is not right for every project: developers with adequate funding who are pursuing a passion project may still choose to proceed without this kind of validation.

Oreshkin says other publishers have been slower to adopt this model because Steam has comparatively less money and competition than mobile, some publishers benefit from strong brands and inbound deal flow, and Steam analytics can be difficult to interpret. He expects more publishers to use creator-based tests as campaign tools become easier to access.

Playtesting: Measure the Product, Not Just the Pitch

Wishlists indicate interest in an idea, but do not prove that the game is enjoyable. After a promising marketing test, Polden’s next step is to playtest as early as possible.

Oreshkin recommends aiming for roughly one hour of gameplay in the test build. He suggests granting access to 2,000–3,000 people, with an expected 1,000–1,500 players and around 500 survey responses. Key measures include median playtime and feedback from a post-session questionnaire.

For the survey, he recommends a simple form—such as Google Forms—that opens in a browser when players quit. He cautions that simply adding a survey button inside the game may not generate many responses. Steam playtest sign-ups can attract people who have already encountered the game and are interested in trying it; their feedback can then guide iteration.

Game Selection and Marketing Fit

Oreshkin identifies three factors that shape a game’s marketing potential:

1. Game idea: A clear, appealing premise, ideally a fresh take on a familiar genre or concept.
2. Setting: A setting with broad appeal. A strong concept can still face limits if its setting attracts a narrow audience.
3. Visual style: An immediately legible and distinctive look that works in short-form content and on a Steam page.

He estimates that a large share of a game’s marketing potential is determined by these choices before a publisher begins promotion. Examples discussed include Total Security: Airport, framed as a “Papers, Please”-style concept about airport security, and familiar-game comparisons such as Hitman reimagined as a co-op game.

A compelling trailer can test a premise even if it is not captured from a playable build, but it should represent a game the team can actually deliver.

Channel and Genre Playbook

• Short-form creators: Particularly useful for games with immediately understandable, visually engaging, or shareable concepts, including some co-op “friendslop,” simulator, and incremental games.
• Organic studio content: If a studio has someone who can make TikToks regularly, Oreshkin recommends testing this for two to three months and evaluating the results.
• Long-form influencers and demos: Often better suited to strategy, roguelike, and larger building games that are less naturally suited to TikTok.
• Steam demos: A demo can re-engage wishlisters through Steam notifications. If those players return and play, their engagement may help the game gain additional visibility.
• Paid performance advertising: Oreshkin says conventional paid acquisition, common in mobile games, generally does not work as well for Steam. Paid traffic sources such as Reddit may be worth considering for some higher-priced strategy games, but he is not confident they work broadly.
• Updates and DLC: For co-op games, ongoing free updates can give influencers a reason to make fresh content and may generate additional sales.

Pricing and Wishlist Benchmarks

Polden generally targets games priced below $10, with many co-op casual games in the $5–$7 range. Some simulators may support prices around $12–$15, depending on their content and expected playtime.

Oreshkin argues that an accessible price can encourage more purchases and help a game meet player expectations. Sales activity may also contribute to further Steam distribution.

Wishlists should not be treated as guaranteed sales from those same users. Oreshkin gives rough benchmarks of about 20% of wishlist count in first-month unit sales and a median of roughly 60–70% over a year. Some games may reach around 100% or more, depending on virality and other factors. These are broad planning estimates, not guarantees.

Publishing Versus Self-Publishing

Oreshkin describes 50/50 as a typical publishing revenue split. He also mentions an 80/20 split while recouping and a 30/70 split when the developer does not need production funding; the interview does not clearly specify the direction of each split.

In his view, a publisher’s most important value is often financing, rather than access to a large set of indispensable Steam marketing tools. Self-publishing can be a good option if a studio has enough development and marketing capital, can assign people to marketing, and is willing to learn the process. He suggests a small team—potentially two people managing marketing—could spend two to three months building its capability.

Marketing is not inherently easy. If a game is difficult to promote, that may point to weak market fit, an unclear premise, or a visual style that does not connect with players.

Company Strategy and Outlook

Polden began without the funding or brand advantages of established publishers. Oreshkin says the company adopted testing as a way to make decisions more carefully. Polden has received investor funding and is growing its slate, while he emphasizes that he wants to be disciplined with investor capital.

The company’s reported position in the interview included just under $8 million in revenue from earlier games and DLC, more than 2.5 million wishlists for upcoming games, and 30-plus games in its portfolio or pipeline.

Near-term areas of focus include simulators, which Oreshkin considers evergreen, and accessible “hyper-casual for Steam” experiences such as incremental, sorting, and simple automation games. Longer term, he wants to build internal studios and a broader ecosystem for indie developers, including education and incubation.

Polden already uses AI in areas such as coding, analytics, and internal documentation. Oreshkin predicts—not as a claim about current performance—that teams of around 10 people could produce AA-scale work within three to five years.

Practical Recommendations

• Validate demand before committing: Test the premise with a short-form campaign and a wishlist target before investing substantially in production.
• Separate market appeal from product quality: Use wishlists to gauge interest in the concept, then use playtime and player surveys to assess the game itself.
• Match channels to genre: Use short-form creators for instantly communicable concepts; consider long-form creators and demos for games that need more explanation.
• Design for discoverability: Prioritize a clear idea, broadly appealing setting, and strong visual identity early.
• Plan independently or seek funding: Self-publish if the studio can fund development and marketing; consider a publisher when financing is the main constraint.
• Treat benchmarks as planning inputs: Wishlist-to-sales ratios vary by game and virality, so use them as rough estimates rather than forecasts.
• Use updates to restart attention: Fresh content can create new influencer coverage and sales opportunities, especially for co-op games.

Presenters and Sources

Host Harry (Hivemind World); guest Kirill Oreshkin, co-founder and CEO of Polden Publishing.