UA Channel Mix Planner

Blended CPI is total spend divided by total installs across every channel — the true average price of a user for your whole program. It is the number your unit economics actually run on: if your value per install is $3.50, individual channels can sit above or below that line as long as the blend stays under it. Judging channels only in isolation leads to a classic mistake — cutting a pricier channel that was feeding volume your cheap channels could never replace, and watching blended CPI rise anyway as you push saturated channels harder.

The blend also frames expansion decisions correctly. A new channel with a $5 CPI looks unaffordable against a $2 target in isolation; against a portfolio blending to $2.80 with headroom, a small allocation may be a rational test. This planner makes that arithmetic visible before the money is spent rather than in the post-mortem.

No channel delivers unlimited installs at its current CPI. Apple Search Ads is bounded by real search volume on your keywords — once you win most relevant auctions, extra budget just goes unspent or chases marginal queries at higher CPT. Social channels scale further but not flatly: pushing past what your best audiences support forces broader targeting, and CPI climbs. A useful cap is the monthly install count at which a channel’s marginal CPI starts exceeding its average — your own spend-versus-installs history shows where that bend sits.

Once caps are honest, allocation logic is straightforward: fill the cheapest channel to its cap, then the next cheapest, until budget runs out — which is what this planner does. The output often surprises: a large budget can exhaust every efficient channel’s capacity and leave money for expensive experiments, while a small budget may fit entirely inside one or two channels, making the others a distraction this month rather than a necessity.