Founders hear the same advice whenever they build on top of another platform, list it in the marketplace, it is free distribution.
The advice sounds rational. The store has search. The platform has trust. The listing is often free. For a small team, it feels like leverage.
The published rules tell a colder story. In most cases, the marketplace is not a first channel. It is a second-order channel that checks whether you already have customers.
Advice and gate
Four official requirements change the distribution math.
- HubSpot requires at least three active unique installs before an app can be listed, and 60 active unique installs plus six months listed before certification, according to its certification requirements.
- Slack will not accept apps installed on fewer than 10 active workspaces, with fewer than 10 weekly active users, according to its Marketplace distribution docs.
- Zapier lets partners publish, but directory feature placement, newsletter spotlight, and the listing Learn More button start at Silver, 50 active users, according to its partner program tiers.
- Chrome caps a new publisher at two published extensions, according to its publishing documentation.
The pattern is simple. The store amplifies proof. It rarely creates proof.
Reference gate list
The table below uses public operator documentation where available. Where a number is not published, it says not published. We did not fill gaps with third-party estimates.
| Marketplace | Fee | Traction gate | Review time | Revenue share | What listing gets |
|---|---|---|---|---|---|
| Shopify App Store | No listing fee found in current app launch docs | No published install minimum for listing. Built for Shopify is a separate quality signal, see Shopify docs. | Not published | 0 percent on the first $1,000,000 USD from Jan 1 2025, then partner keeps 85 percent, plus 2.9 percent processing fee, per Shopify revenue share. | A store page, search eligibility, ads eligibility, and ranking benefit if quality standards are met. |
| Slack Marketplace | Not published | 10 active workspaces and 10 weekly active users, per Slack docs. | Up to 10 business days preliminary review, then up to 10 weeks functional review. | Not published | Marketplace distribution after review, not cold-start access. |
| HubSpot App Marketplace | No listing fee, no revenue sharing, per HubSpot docs. | Three active unique installs to list. 60 active unique installs and six months listed to certify. | Not published | None, per HubSpot. | Listing first, certification later, one app submitted at a time. |
| Zapier | Free to publish, per Zapier publishing docs. | No published gate to publish. Visibility benefits start at Silver, 50 active users, then Gold at 350, Platinum at 3,000. | Not published | Not published | Public integration listing. Meaningful promotion starts with active user thresholds. |
| Chrome Web Store | Registration fee amount not published in Google terms. The agreement says the amount is determined by Google, see Chrome Web Store terms. | New publishers are limited to two published extensions, per Chrome publishing docs. | Not published as a fixed service level | Not published in current cited docs | A public listing, with ranking influenced by ratings and usage statistics. |
| Atlassian Marketplace | Partner registration fee not published | No published install minimum found | Not published | Partner share is published, Forge 84 percent, Connect 80 percent, Data Center 75 percent, changing Oct 1 2026 to 83, 75, and 75, per Atlassian pricing docs. | A searchable marketplace listing, with unusually transparent ranking signals. |
| Salesforce AppExchange | Paid app security review and annual listing fees are commonly stated as $2,700 and $150, but we treat those as medium confidence because the older fee page is not stable. Salesforce explains the contractual model in Trailhead. | No published install minimum found | Four to six weeks is the commonly stated review window, medium confidence | Salesforce takes roughly 15 percent, per its partner model materials. | Enterprise procurement trust, security review, and AppExchange discovery. |
| Notion | Not published | No published install minimum found in public integration materials | Not published | Not published | Integration visibility, subject to Notion review and public integration requirements. |
The brief calls for nine marketplaces. The public source set supports eight with enough confidence to publish. We are not inventing a ninth row to make the table look complete.
Is it free?
Sometimes. But free listing and low effective cost are different things.
Shopify is the clearest example. Its canonical revenue share page says developers keep 100 percent of the first $1,000,000 USD in gross app revenue earned from January 1 2025, then 85 percent above that. Billing still carries a 2.9 percent processing fee.
A separate Shopify app review page currently uses older 20 percent to 15 percent framing. Treat the revenue share page as authoritative, and treat the conflict as a reminder to check operator docs before modeling margin.
HubSpot is simpler. Its developer docs state there is no listing fee and no revenue sharing. Atlassian publishes the partner share rather than the marketplace cut. Salesforce is the opposite pattern, where the review and listing economics matter before the first customer is acquired.
The cheapest marketplace to enter is not always the cheapest marketplace to grow through.
Store ranking signals
Most marketplaces do not publish their ranking formulas. Three give enough signal to guide decisions.
Chrome is unusually plain. Its discovery documentation says ranking considers user ratings and usage statistics, including downloads versus uninstalls over time. It also states featuring cannot be purchased.
Atlassian is the most transparent. Its ranking documentation names keyword relevance, semantic match, and engagement score. It also names the fields those systems read, including app name, tagline, brief description, developer name, ratings, review count, average rating, install count, and support details.
Shopify does not publish a complete formula. It does say Built for Shopify apps rank higher, that off-platform traffic can support discoverability, and that keyword overuse can reduce discoverability in its marketing guidance.
We found no comparable ranking documentation from Slack, HubSpot, Zapier, Salesforce, or Notion. Anyone giving precise algorithm advice for those stores is guessing.
Install economics
The most useful marketplace data often comes from founders, not operators.
In July 2025, ezApps published a first-year Shopify report with 1,171 total installs, 584 uninstalls, 587 active merchants, and $735.66 in revenue. That is roughly $1.25 per active merchant for the year, using the founder's own numbers from his revenue report.
The more important line is attribution. In a companion post, the founder wrote that one detailed answer in the Shopify Community became his number one traffic driver for four months, beating app store search, according to his growth breakdown.
The first channel was not the marketplace. It was a public answer to a real customer problem.
There is a countercase. DelightChat reached more than 20,000 active merchants, more than 1,500 paying customers, and passed $25,000 MRR with a two-person team, according to Preetam Nath's case study. The same post warns that depending on one marketplace is dangerous, because a competitor overtook their ranking in under three months.
Both outcomes are real. The difference is whether the marketplace is treated as the first channel, or as leverage after demand exists.
Are listings searchable?
Yes, but the answer depends on the search layer.
On 16 Sep 2026, logged out in the US, we observed marketplace pages ranking at or near the top of Google for three commercial tests. apps.shopify.com held organic position one for best shopify app for abandoned cart recovery. chromewebstore.google.com held positions one and two for best chrome extension for full page screenshot. A slack.com marketplace page and a Zapier app-pair page both appeared on page one for how to connect airtable to slack automatically.
Google rendered no AI Overview for any of those three observed searches. This is a point-in-time observation, not a claim about all marketplace SERPs.
Perplexity behaved differently the same day. For the Chrome screenshot extension question, it cited three Chrome Web Store listing pages directly. For the Shopify abandoned cart question, it cited zero apps.shopify.com pages and built the answer from vendor and affiliate roundup blogs.
The practical read is narrow, but useful. In some categories, the listing is the citable asset. In others, the roundup post is. A founder may need to be in someone else's comparison page before the marketplace listing compounds.
Where start first?
Sequence the work by the gate you can clear today, not by the marketplace logo you want on the website.
- Before installs, use surfaces with no traction gate. Publish the integration, if allowed, but spend more energy on community answers, docs, comparison pages, and problem-specific search assets. This is the channel that got ezApps moving.
- Around ten customers, unlock stores that require real usage. Slack becomes plausible at 10 active workspaces and 10 weekly active users. HubSpot listing becomes plausible once three active unique installs exist.
- Around fifty users, pursue amplification. Zapier Silver starts at 50 active users. HubSpot certification needs 60 active unique installs and six months listed. Shopify quality programs and ranking signals become more meaningful once retention exists.
- At scale, manage marketplace risk. Built for Shopify status, Atlassian engagement score, Chrome ratings, and third-party comparison coverage all become distribution infrastructure. They are not one-time launch tasks.
This is also where search visibility matters. A marketplace listing has to rank twice, inside the store and on the open web. For many early teams, that work belongs in the same operating system as content, community, and partner pages, not in a separate launch checklist. Better Marketing frames this as search visibility, because the page that wins is often not the page you own.
What remains unknown
The public record is incomplete.
- No marketplace operator except Shopify publishes meaningful install or traffic data for small apps.
- Shopify's statement that 60 percent of App Store installs came from search is from a February 2020 Shopify Partners post, which makes it useful historical context, not current market sizing.
- Shopify's claim that Built for Shopify apps see an average 49 percent increase in new installs within 14 days comes from an operator-published 2025 post, so it should be read as a marketing claim with a number attached.
- Google publishes no aggregate Chrome Web Store traffic data that would let a founder model expected installs.
- Zapier publishes no referral traffic data for partners.
- We found no Chrome extension or Zapier partner publishing a clean signup breakdown attributed to marketplace search.
- Third-party estimate vendors, including Store Leads, StoreCensus, StackDex, and DebugBear, may publish useful estimates. We did not verify them, and we are not citing them here.
This uncertainty is the point. A founder should not spend a sprint assuming a marketplace will create demand when the operator does not publish the demand curve.
Takeaway
Check the published traction gate before building the listing plan.
If you cannot clear the gate, the marketplace is not your first channel. The first channel is usually a public answer, a workflow guide, a community thread, a partner page, or a comparison article that proves demand exists.
Then the marketplace can do its job. It can verify trust, capture search demand, and amplify traction you already earned.
That is the quiet correction founders need. The store is not free distribution. It is rented distribution with an entrance exam.