← Back to blog
Noob Games Lab Blog

Before You Spend More on Marketing, Find Where Players Drop Off

By Max G. 6 min read
  • indie
  • design
  • tech
A magnifying glass focuses on a break in a blue pipe carrying a stream of red and blue circles.

When you have spent weeks promoting a game and the numbers barely move, it is tempting to buy more traffic or disappear into another month of polish. Both can feel productive. Neither tells you what went wrong. Before committing to either, find out whether people misunderstand the game, struggle to start it, or simply are not the audience you meant to reach.

The numbers raise questions, not a verdict

In a Reddit discussion, the developer of Arcanum Solitarius reports more than 6,000 store-page visits, 50 wishlists, roughly 1,100 demo downloads and only 25 people running the demo over a couple of months. They are wondering whether to proceed with Steam Next Fest or spend more time on the game. These are the developer's reported figures; we have not seen the underlying reports.

Dividing 50 by 6,000 gives about 0.8%. That is a rough ratio of the reported totals, not a verified visitor conversion rate. We do not know whether the counts cover matching periods, include repeat visits, or describe the same people. The gap between downloads and launches also deserves a check, but it does not tell us whether players changed their minds, saved the demo for later, or encountered a technical problem.

Check separate decisions, not one neat funnel

A player can wishlist a game without trying it, play a demo without wishlisting, or return long after their first visit. Treating every total as the next step in one journey creates certainty the data cannot support. For this case, we would investigate the page and the demo separately, then look for evidence connecting them.

These are the questions worth answering:

  • Impression to visit: when people actually see a clickable capsule, does it give them a reason to investigate?
  • Visit to wishlist: can the intended audience understand the game and find a reason to want it?
  • Download to play: what does the download count represent, and can a new player reach a working session?

The post gives no impression count or source breakdown, so it cannot establish whether the capsule earns clicks. It does establish that visits are arriving. That makes “nobody is seeing the game” an incomplete explanation, not proof that enough of the right people have seen it.

Turn criticism into something you can test

Several commenters object to the game's AI-generated visuals; one points to specific card-layout and readability problems. That is feedback worth investigating, not evidence that replacing every asset will solve demand. Ask target players what looks unclear, unfinished or inconsistent, and whether it stops them wanting to play. A cramped card label suggests a focused readability fix. A broader loss of trust may call for a different visual approach. Neither finding automatically justifies an expensive art overhaul, and making a page clearer will not rescue a premise that those players do not want.

Compare relevant audiences before chasing a benchmark

A visit from a fan of similar games and a visit from someone casually browsing need not mean the same thing. Start with Steamworks traffic reports to see which sources bring people to the page. Compare your own results over matching periods, noting changes in audience and promotion. Avoid grading the page against a universal percentage when you do not know how the comparison was measured.

For future external promotion, use separate UTM-tagged links for each campaign. Steamworks reports attributed wishlist additions from tracked visits, but switching apps or devices can leave actions unattributed. Use that information alongside player feedback, not as a complete account of every visitor. If a relevant community responds better than broad traffic, test the targeting further before deciding that the whole page needs rebuilding.

Run a small audit that leads to a specific change

You do not need a large campaign to begin. We would start with these five checks. A small round of player sessions can reveal problems to fix; it cannot establish a reliable market-wide conversion rate.

  1. Make the reports comparable. Use the same dates for the base game's traffic and wishlist additions. Check what the demo report labels as downloads, licenses or players; do not quietly substitute one for another. Note which app each number belongs to. If the apparent gap comes from incompatible counts, correct the comparison before changing the game.
  2. Test the page without explaining it. Show it to a few people who already enjoy similar games. Ask what they expect to do, what makes this game different, and what would stop them trying it. Let them answer before you defend the design. Repeated misunderstandings suggest a clearer description, gameplay footage or screenshots; clear understanding without interest calls for a closer look at the pitch and audience.
  3. Try the public demo on a fresh setup. Have someone outside your development environment get it through Steam and launch it on a supported machine. Watch the actual path to the first playable moment. A missing dependency, confusing menu or crash is a concrete fix. One successful test does not prove every installation works, so record the setup and investigate reported failures too.
  4. Watch where play becomes confusing. Let a new player start without coaching. Note the first unclear goal, unreadable card or moment when they cannot tell whether an action worked. Afterwards, ask what they expected and whether they want another session. Fix repeated obstacles before adding more content. If they understand the game but do not enjoy its central decisions, more tutorial text is unlikely to be the answer.
  5. Change one thing and check again. Pick the strongest finding: a clearer pitch, a readability fix or a launch repair. Record the change and retest that problem. Compare traffic results only over reasonably similar periods and audiences, allowing for attribution delays. Treat a handful of extra wishlists as a reason to keep testing, not proof that the fix caused a lasting improvement.

Choose Next Fest for a reason, not out of panic

Under Valve's current rules, a title can participate in only one Steam Next Fest. That makes the timing decision more important than “try it and see.” Learning from the event is a legitimate goal, but basic launch and comprehension checks are better done beforehand. If those checks expose problems you cannot resolve in time, consider a later eligible edition that fits your release plan rather than assuming you can repeat the event.

If target players understand the promise, the demo starts reliably in your tests, and you can act on feedback, participating may be reasonable even with a small wishlist count. Set a question you want the event to answer, such as whether fans of similar games want to keep playing after the opening. We cannot choose the right edition for this developer from the post alone. Our recommendation is narrower: fix the strongest observed obstacle, check again, and spend on reaching more players when you can explain who you are trying to reach and why.