# GHAN vs. Google Ads for desktop app installs

> Google Ads sells the highest-intent traffic there is, at a price set by whichever bidder in your keyword has the most lifetime value per customer. For a EUR 29 one-time desktop licence, that bidder is almost never you — which is an arithmetic problem, not a creative one.

Source: https://ghan.io/compare/google-ads/  
Published: 2026-09-22 · Updated: 2026-09-22  
Publisher: GHAN — the audited cross-promotion network for desktop apps (https://ghan.io)

---

## The arithmetic problem

A Google Ads click in a desktop software category is priced by the bidder with the most to gain from it. In most categories that is a subscription product.

| | One-time licence | Subscription SaaS |
|---|---|---|
| Price | EUR 29, once | EUR 29 per month |
| Gross lifetime value | ~EUR 29 | ~EUR 350 at 12 months |
| Affordable cost per customer at 30% margin | ~EUR 9 | ~EUR 105 |
| Clicks per customer at 3% conversion | 33 | 33 |
| Affordable cost per click | **EUR 0.27** | **EUR 3.18** |

Both products are bidding on the same keyword. One of them can pay almost twelve times more for the identical click. No amount of ad copy testing closes a twelve-fold gap, and no landing page optimisation invents lifetime value that the pricing model does not produce.

This is why desktop utilities have quietly stopped running search: not because the channel is bad, but because the channel is an auction and they brought the wrong balance sheet.

## What GHAN changes

GHAN does not abolish the auction — [it has one](/bidding/), because a price that never responds to demand cannot clear a market or reward the apps with the best inventory. What GHAN changes is what the auction sits on top of.

In a search auction the clearing price *is* the price. There is no floor to fall back to, so when a better-funded competitor enters your keyword you are not outbid on some impressions, you are priced out of the channel. GHAN publishes a floor that never rises, reserves 30 percent of every category at that floor, and auctions only priority above it. And you bid per **cleared install** rather than per click, so a lost auction costs nothing and a won one costs nothing unless the person is still using your software two days later.

On GHAN you are not paying cash you had to find; you are paying with attention you already have. An app that serves placements earns credits, and credits buy installs at the same published rate card everyone else pays.

| | Google Ads | GHAN |
|---|---|---|
| Price floor | None — the clearing price is the price | **Published, permanent, always available** |
| Price above the floor | Auction, by highest-LTV bidder | Second-price auction for priority only |
| Can you be priced out entirely | **Yes** | No — 30% of each category reserved at floor |
| You pay for | A click | An install still in use at 48 hours |
| Charged when install fails | Yes | No — automatically refunded |
| Audience | People searching for your category | People installing desktop software right now |
| Creative | Yours, unlimited | Templated: icon, name, one line |
| Can you pay in non-cash | No | Yes — earn credits by serving |
| Attribution | Platform-reported | Two independent signatures, auditable |
| Minimum spend | Effectively yes | None |

## The honest version of the comparison

Search wins on intent. Somebody typing "best clipboard manager for mac" is more valuable than somebody who just finished installing a screen recorder, and they always will be.

GHAN wins on three things search cannot offer a small desktop app: a price that does not move when a better-funded competitor enters your keyword, a unit of billing that is already net of churn, and the ability to pay in inventory rather than in money.

The practical answer for most apps is both — run search where it pays back, and serve placements to fund the volume search cannot reach.

## Measuring them against each other

Compare on cost per **retained** user, not cost per install, and use the same retention window for both. GHAN's rate card is already a retained price because installs that do not survive 48 hours are refunded. To make search comparable, take your search spend, divide by the installs still active at 48 hours, and compare that number to the [rate card](/pricing/) for your category.

Most apps doing this exercise for the first time find their real search cost per retained user is two to four times their reported cost per install.

## Questions people ask about this

### Should a desktop app stop running Google Ads?

No, if it is paying back. Search captures people who are actively looking for what you make, which is the best traffic in existence, and nothing else replicates that intent. The question is not whether search works but whether it works at your price point, and for one-time licences under roughly EUR 40 the answer is frequently no once you account for the share of clicks that never install and the share of installs that never stay.

### Why are desktop app keywords so expensive?

Because the price is set by the highest bidder in the auction, not by the value of the traffic to you. A subscription tool billing EUR 30 a month can profitably pay fifty times what a EUR 29 one-time licence can for the identical click, and it is in the same keyword. You are not competing on relevance; you are competing on lifetime value, and you brought a one-time payment to a recurring-revenue fight.

### How does cost per retained user differ from cost per install?

Cost per install divides spend by installs. Cost per retained user divides the same spend by the installs that were still being used days later. The second number is often two to four times the first, and it is the only one that corresponds to revenue. GHAN's pricing collapses the gap by refusing to bill for installs that do not survive 48 hours, so the price you are quoted is already the retained price.

### Can GHAN replace paid search entirely?

For most apps, no, and claiming otherwise would be silly. Cross-promotion reaches people installing desktop software right now who were not searching for you, which is a genuinely different audience from search intent. The realistic position is that cross-promotion is the channel whose economics do not depend on your price point, so it is the one that works while search does not.