What it costs to acquire a mobile app user in Singapore, and the budget rule Google's own page does not state

User acquisition cost for a mobile app in Singapore is not a price you look up, it is a ceiling you set and then defend. On both of the two platforms that matter, the published mechanism is that you nominate the figure: Apple Ads Basic states "Pay only for installs at a price you set" and that "We suggest a max cost-per-install, or choose your own", read on ads.apple.com/app-store/basic on 28 October 2026, and Google App campaigns are bid with a target cost per install. So the honest answer to "what does an install cost in Singapore" is a planning range to start from and an arithmetic to replace it with.

The Singapore planning ranges that are published, all read on 28 October 2026, come from one Singapore agency's own guides and they do not fully agree with each other. Its Google App Campaigns guide states that "In Singapore, typical CPIs range from SGD 1.50 to SGD 8.00 depending on category and competition". Its Google Ads for Apps page narrows that to "CPIs for quality users typically range from SGD 2 to SGD 8". Its app marketing guide splits by category instead: "Gaming apps typically see CPI from S$1.50 to S$5, whilst fintech and e-commerce apps range from S$3 to S$12. B2B and niche apps can exceed S$15 per install." Three pages on one site, three different shapes of the same answer, which tells you the figure is a category question rather than a country question.

The rule everybody repeats, and what Google's own page says

The most quoted number in this subject is a budget multiple. Both Singapore pages above state it the same way: "Google recommends a daily budget of at least 50 times your target cost per install (CPI) for install campaigns, or 10 times your target cost per action (CPA) for engagement campaigns", and both convert it, giving "a minimum daily budget of SGD 100 to SGD 400" for a Singapore app at a target of SGD 2 to SGD 8.

That conversion is arithmetic you can check and it is correct against its own inputs. The attribution is the part to be careful with. Google's own About App campaigns page at support.google.com/google-ads/answer/6247380, read on 28 October 2026, states no recommended daily budget of any kind and no multiple of target CPI. Its only budget sentence is that an advertiser needs to "provide some text, a starting bid and budget". So the fifty times rule is widely repeated Singapore agency guidance that reads as a platform requirement, and on the platform's own overview page it is not there.

Use it anyway, as the sensible operating rule it is, and know what it is. A bidding system needs enough daily conversions to learn from, and a daily budget of two or three installs produces a campaign that never leaves its learning phase. Fifty times the target is a defensible floor. It is not a published minimum, and nobody at Google will enforce it.

Work the ceiling backwards, which is the only figure that matters

Your target CPI is whatever your app earns per user, minus the store's cut, minus the margin you want to keep. The store cut is published and it is the line most Singapore plans leave out.

Google Play's service fee page, read on 28 October 2026, sets its standard rates for Australia, the EEA, Japan, the United Kingdom and the United States, and a separate rate for the remaining markets, which is where a Singapore developer selling to Singapore users sits. For those remaining markets the page states "15% for the first $1M (USD) revenue earned by the developer each year" and 30 per cent above that threshold, with auto renewing subscriptions at 15 per cent regardless of annual revenue.

So the arithmetic for a subscription app looks like this, and the numbers are yours rather than ours. Take the average revenue a paying subscriber produces before they churn. Take 15 per cent off it for the store. Multiply by your free to paid conversion rate, because most installs never pay. What is left is the absolute maximum an install can cost you, and your target CPI should sit well under it so there is a business rather than a break even. The Singapore agency page states the same logic more briefly for the four app models: for in app purchase apps the "Target CPI should be below the average revenue per user (ARPU) over a meaningful time period", for ad supported apps below average ad revenue per user, and for ecommerce apps below the average profit margin per customer acquired through the app.

One worked illustration, using the figures as published rather than inventing any. If an app's paying subscriber is worth SGD 50 over their lifetime, the same guide suggests "a tCPA of SGD 15 to SGD 20 might be sustainable" for an in app action. Note that is a cost per action, not a cost per install. If one in ten installs becomes that action, the install ceiling is a tenth of it, and that is the moment a SGD 8 install stops looking cheap.

What each platform publishes, and what it does not

Two things in that table are worth saying out loud. Apple's figure for its own starting credit is denominated in US dollars even though Singapore is a supported region, which is the same pattern as almost every tool a Singapore buyer subscribes to. And the Singapore agency guidance names three channels rather than two, stating that "most successful app marketers in Singapore run campaigns across all three simultaneously" across Google, Meta and TikTok, with TikTok's strength put at users aged 18 to 35.

Why cost per install is the wrong number to optimise

The Singapore app marketing guide puts the case bluntly, and the sentence is worth quoting because it is the whole argument against buying cheap installs: "The average mobile app loses 77% of its daily active users within the first three days after install." Whether your app matches that or beats it, the implication holds. An install that churns on day two has consumed the acquisition budget and produced nothing, so a cheaper install is only better if it retains at the same rate.

That gives you the four numbers to put beside CPI in any Singapore report: registrations completed, day 7 retention, first purchase or first meaningful action, and revenue per cohort. Cohort them by acquisition channel, because the point of the exercise is to find out which channel brings people who stay. The same guide makes the comparison that settles most budget arguments: "A S$10 install that generates S$50 in revenue is far more valuable than a S$2 install that churns immediately."

Two Singapore specific checks before you scale

Separate the market. The local guidance is explicit that if you target Singapore and other ASEAN markets you should "create separate campaigns", because "User behaviour, CPIs, and competitive dynamics differ significantly across markets". A blended Singapore and Malaysia install figure hides which one is carrying the result, and in a market this small the blend usually hides Singapore.

Do not confuse this number with your customer acquisition cost. An install cost is a media figure. Customer acquisition cost in Singapore includes the people doing the work, which is usually the largest line, and it behaves differently as you grow. Our page on what it costs to acquire a customer in Singapore sets out the four inputs that move that figure, and the two are reported separately in any plan worth approving.

The cheapest install is the one you do not buy

Before any of this, the store listing itself is the lever with no media cost attached. The local guidance makes the point that in Singapore "a well-optimised listing can reduce your dependency on paid acquisition and lower overall cost per install", and that the app title is the most influential ranking element inside the store. It also makes a genuinely local observation about keyword choice: Singapore app searches include Singlish and bilingual patterns, so a food delivery app might reasonably target "dabao near me" alongside the English phrasing. Google Play also offers native A to B testing of store listings, and a better store conversion rate reduces your effective install cost on every campaign at once.

If that is the work you want done rather than the media bought, our page on what an app store optimisation engagement actually covers sets out what the stores let you test and how to judge the result.

Questions about app install and user acquisition costs in Singapore

What is a realistic cost per install in Singapore?

On the Singapore figures published on 28 October 2026, the range quoted for planning is SGD 1.50 to SGD 8.00 across categories, narrowing to SGD 2 to SGD 8 for users described as quality installs, with gaming at the low end and finance, business and B2B apps above it. Use the range to set a first bid and replace it within a fortnight with your own measured figure, because your own number is the only one that reflects your creative, your listing and your category.

Is there a minimum budget for app install campaigns?

No platform minimum was published on Google's own App campaigns overview page when we read it on 28 October 2026. The working floor repeated by Singapore agencies is fifty times your target install cost as a daily budget, which converts to roughly SGD 100 to SGD 400 a day per campaign at Singapore install prices. One local page puts a monthly figure on the same thing, at SGD 3,000 to SGD 5,000 a month per campaign to generate enough conversions to decide anything.

Should I launch paid acquisition on iOS or Android first in Singapore?

Decide it on where your paying users are rather than on install price, because the store cut differs by market tier and the monetisation model differs by audience. Google Play's published rate for markets outside Australia, the EEA, Japan, the United Kingdom and the United States is 15 per cent on the first USD 1,000,000 of annual developer revenue, which is a meaningful advantage for an early stage Singapore app against the 30 per cent band above it. Run the arithmetic on both before you pick, and pick one to start so the data is clean.

Why does my install cost rise as I spend more?

Because Singapore is a small auction. Each additional dollar reaches a slightly less interested person than the last one, so the marginal install is dearer than the average install. That is why a monthly average is a poor control figure and a weekly cohort is a good one, and why scaling in steps of 15 to 20 per cent against a target rather than doubling a budget is the approach the local guidance recommends.

Does a Singapore install cost include GST?

Advertising bought from an overseas platform is an imported service, so the treatment depends on your own GST registration rather than on the ad platform's billing page. Settle which regime you are in before you compare a platform invoice against an agency quote, and ask for the figure both ways.

The install you never see reported

Every number on this page comes from a system that counts a tap and then a download. There is a route into an app install now that produces no tap on any of those dashboards: somebody asks ChatGPT, Gemini, Perplexity, Claude or Google AI Overviews which app to use for the thing they want to do in Singapore, is read two or three names, and goes to the store and searches for one of them by name. In your reporting that lands as organic store traffic with no source, and in the assistant's answer it was a recommendation somebody else's app earned. A dated reading of what those five surfaces name in your app's category is a small piece of work with no media budget attached, and the first one at aiknowsus.com needs no card and no SDK.