You License HTML5 Games by the Thousand. Your Homepage Has Shelf Space for Forty.

When you license HTML5 games as a catalogue, the licence is the easy part. Surfacing a thousand titles to real players is the constraint nobody budgets for.

Every portal operator I have watched go live has run the same sequence. Negotiate hard on catalogue size. Sign. Ingest. Then spend the next six months discovering that the homepage introduces roughly the same forty games it introduced in week one, and the other nine hundred sit in a database being paid for.

This is not a licensing failure. The licence delivered exactly what was agreed. It is a merchandising failure, and it is structural — the number of titles you can license scales freely, and the number of titles a player will ever be shown does not. Getting the second number up is the actual work of running a portal, and almost nobody prices it into the launch plan.

🧮 Two Screenfuls Is the Whole Storefront

Nielsen Norman Group's eye-tracking work on scrolling is the clearest data on how much of a page anyone actually looks at. In their study, users spent about 57% of their page-viewing time above the fold, and 74% of the viewing time was spent in the first two screenfuls, up to 2160px. The remaining 26% was spread thinly down the rest of the page.

Turn that into tiles. This is arithmetic from stated assumptions, not a measured statistic, so check it against your own grid: a desktop games portal typically runs five or six thumbnails across, and a tile with its label occupies somewhere around 200–280px of vertical space once you include the gap. Two screenfuls — 2160px — is therefore roughly seven or eight rows. Six across by seven rows is 42 tiles. Call it forty.

On mobile it is worse. Two or three across, shorter screenfuls, and the same sharp attention decay. Your mobile homepage introduces something closer to twenty games before three-quarters of the available attention is gone.

Forty is the shelf. It does not grow when the catalogue does.

📉 Every Title You Add Competes for the Same Forty Slots

The consequence is that catalogue growth past a certain point is zero-sum against your own inventory. Title 900 does not get a new slot. It has to take one from a title that is currently earning it.

Most portals never let that happen, because most portals rank the grid by plays. A title that has been on the homepage for a year has a year of accumulated plays. A title ingested last Tuesday has none. Ranking by plays is therefore ranking by how long something has already been on the homepage, dressed up as a performance metric. The incumbent set entrenches itself, and it does so faster the more traffic you have.

There is a second-order effect worth naming. If your top forty never turns over, your returning-player experience is a static shop window. The reason to license depth is that a player on visit six sees something they have not seen; a frozen grid throws that away and leaves you paying catalogue prices for a landing page.

🗂️ Taxonomy Is the Only Thing That Multiplies Shelf Space

You cannot make the homepage taller in any way that survives contact with the attention data. What you can do is have more storefronts. Every category page is its own two screenfuls, with its own forty slots and its own entry traffic from search and internal links.

Thirty genuine category pages that each get real traffic is 1,200 slots. That is the whole shape of the fix, and it is why taxonomy work returns more than any further licensing spend once you are past a few hundred titles.

"Genuine" is doing work in that sentence. A category is genuine if a player would plausibly type it, if it holds enough titles to fill two screenfuls, and if it is not a near-duplicate of another one. Twelve overlapping puzzle categories with nine games each is not thirty storefronts; it is one storefront with a broken filing system.

Tag on axes players actually choose by

Genre is the obvious axis and the least differentiating, because every competitor has it. The axes that build distinct, defensible surfaces are the operational ones:

• Control scheme — one-button, mouse-only, keyboard-required, touch-native. This is the single most useful tag for a portal that serves both desktop and mobile, and almost nobody stores it.
• Orientation — portrait, landscape, or genuinely responsive. Determines what you can put in a mobile-first collection at all.
• Session length — a sub-two-minute title and a twenty-minute progression game belong on different surfaces and are measured differently.
• Players — single, local two-player, online multiplayer.
• Age suitability — necessary for anything you sell into schools, kids' portals or carrier decks.
• Art style and theme — cartoon, pixel, realistic, sports, cooking, driving. Weak on its own, strong in combination.

For scale reference: published figures for Poki, one of the largest web games platforms, disagree — its Google Play listing describes 1500+ free online games, while other coverage of the platform refers to a curated library of 1,000-plus titles. Take the range rather than either number. Either way, a platform operating at that scale is not running a five-figure library. Catalogue size in the low thousands is normal at the top of the market, which means depth is rarely the thing separating your portal from theirs. Presentation is.

🔍 On-Site Search Is Your Second Storefront, and Most Are Bad

Search is the one surface where a player tells you exactly what they want, and it is routinely the worst-built part of a games portal. Baymard Institute's 2026 benchmark found that 56% of sites have "mediocre or worse" Search UX. Their autocomplete research is starker still: the feature is provided at 80% of e-commerce sites, yet only 19% achieve the highest performance rating.

Game search has failure modes that general e-commerce does not:

• Misspelt titles. Players half-remember names. Without fuzzy matching, a one-character error returns zero results and the session ends.
• Character and brand names that appear in a title's description but not its name.
• Genre words typed as searches — "racing", "dress up", "bubble" — which should land on a category page, not a thin result list.
• "Games like X." Nothing in a title-only index can answer this. A tag index can.

If your search matches only the title field, you are failing three of those four. Indexing descriptions and tags alongside titles is a day of work and it is usually the highest-yield day available to a portal operator.

Log your zero-result queries and read them weekly. That log is the only free, direct, unfiltered statement of demand you will ever get, and it tells you what to license next far more reliably than a genre chart.

🏷️ The Metadata Is a Licence Deliverable, Not a Content Project

Here is where merchandising becomes a procurement question. Everything above depends on structured data about each title — and if the licence does not specify it, you will be writing it yourself, 1,049 times.

Put these in the deliverables schedule, by name:

• Thumbnails in more than one aspect ratio, at a stated minimum resolution — a square-only asset pack makes a hero row impossible.
• A short description and a long description, per title.
• Genre plus free-form tags, and the operational axes above where the licensor holds them.
• Supported languages per title, and whether the in-game text is localised or only the wrapper.
• Orientation, minimum viewport, and control scheme.
• A machine-readable feed — the difference between an afternoon and a quarter is whether this arrives as a structured export or a spreadsheet.

One clause matters more than the asset list: the right to rewrite the copy, not merely to display it. Every licensee of a given catalogue receives the same descriptions. If all of you publish them verbatim, you are competing for the same search results with identical text. Google's own crawling documentation puts the instruction bluntly — "consolidate duplicate content to focus crawling on unique content rather than unique URLs" — and the version of your page that survives consolidation may not be yours.

Worth knowing, because SEO vendors will try to sell you the opposite: crawl budget itself is almost certainly not your problem. Google scopes that guide to large sites with 1 million+ unique pages, or medium and larger sites with 10,000+ unique pages and very rapidly changing content. A portal with a few thousand pages is neither. Duplicate text is your problem. Crawl budget is not.

🔁 Rotation: Give Every New Title a Fair Test

If the grid ranks by lifetime plays, nothing new can ever win. The fix is to reserve a fixed share of the shelf for exploration and to judge new titles on a rate rather than a total.

A workable version, which you can run without any cooperation from the licensor:

1. Reserve 15–20% of homepage and category slots as exploration inventory.
2. Give each new title a fixed impression budget in those slots — a few thousand impressions is usually enough to separate signal from noise.
3. Judge on click-through per impression, not plays. Plays measure placement; CTR per impression measures the thumbnail and the title.
4. Promote the top performers into the ranked grid, return the rest to the long tail, and re-test after any thumbnail change.

Be honest about the limit of this. Click-through tells you whether the tile sold the game. It tells you nothing about whether the game held anyone, because a cross-origin embedded build does not report that back. Rotation optimises the shop window; it cannot grade the product. Keep the two questions separate in the meeting.

🚫 Five Ways Operators Waste a Catalogue They Paid For

• Negotiating title count and nothing else. A licence for 1,000 titles with no metadata schedule is a licence for 1,000 titles you will spend a quarter making presentable.
• One flat "All Games" page with infinite scroll. It looks like it exposes everything and exposes the first two screenfuls, same as any other page — while generating no category surfaces for search.
• Ranking permanently by lifetime plays. Guarantees the launch-week grid is still the grid a year later.
• Publishing the licensor's descriptions verbatim. You inherit the duplicate-content position of every other licensee.
• Ingesting the whole catalogue on day one. A staged release gives you a supply of genuinely new titles to merchandise for months, at no extra licence cost.

🎯 What You Get When You License HTML5 Games From a Direct Licensor

Forestry Games has licensed games since 2017 and holds a catalogue of 1,049 titles spanning HTML5 and Android APK builds, with titles also published on Google Play and the App Store. A licence conversation covers scope — territory, platforms, term, exclusivity — and delivery: HTML5 builds for embedding or self-hosting, APK builds where the use case is Android, source where applicable, branding, and hosting options if you would rather not run the files yourself.

For a portal operator the practical point is that catalogue depth and format breadth sit in one conversation rather than several. You can size the initial ingest against the surfaces you can actually merchandise, hold the rest in reserve for staged release, and add the Android format later without re-opening a separate negotiation. Start by working through the catalogue against your own category plan, or read how the formats differ before you commit to a shelf design at license HTML5 games and APK game licensing.

🧸 Licensing Branded Games for Campaigns, Portals and Events

Branded titles are the exception to everything above, because a recognised character does the merchandising work that a thumbnail otherwise has to do alone. Forestry Games works with branded IP and has brand partnerships including Disney, Nickelodeon, Cartoon Network and Warner Bros, and businesses can license branded game content through it for campaigns, portals, events and apps.

A branded licence is scoped differently from a generic catalogue licence — approvals, territories and permitted surfaces are all narrower, and that shapes where a branded title can sit on your portal. If a branded row is part of your plan, raise it at scope stage rather than after the taxonomy is built. Ask for a licence scope, or request a white-label game portal demo to see how a catalogue behaves inside a working front end.

🧭 What to Do With Your Top Fifty This Week

Export your top fifty game pages by plays and put a second column beside it: the date each title first appeared on the homepage. If the two columns correlate tightly, you are not looking at a ranking of your games. You are looking at a ranking of your merchandising decisions from launch week, and the licence fee on the other nine hundred titles is buying an archive.

Fix it in this order, because the order matters: tag the catalogue on the operational axes first, build category pages against the tags, index descriptions and tags in search, then turn on rotation. Taxonomy before search before rotation — rotation without categories just shuffles the same forty slots.

And when the shelf is finally wider than the catalogue can fill, that is the moment to license more titles, not before. If you are at that point now, browse the catalogue and ask for a licence scope built around the surfaces you have, rather than a title count that sounds impressive in a board pack.