Case studies are useless. Every mobile-app vendor (us included) has a "Brand X grew 145% in six months" story. Some are real, some are cherry-picked, none of them help you figure out what your own store would actually do on mobile. This article shows you the math we use internally when a Shopify brand asks us whether the build is worth it for them specifically. Bring your own analytics tab; you will need it.
We will walk through five inputs — mobile-web baseline, install rate, install-to-purchase conversion, repeat-purchase lift, and push-driven incremental revenue — then layer on the build cost and platform fee to get a payback period. The numbers we plug in are deliberately conservative; if you are bullish, lift them. The model still works.
Input 1: your mobile-web baseline
Open Shopify Analytics. Filter to mobile devices for the last 90 days. Note three numbers: total mobile sessions, total mobile revenue, and mobile conversion rate. These are your baseline. Everything in the model is a multiplier on these numbers, so it does not matter whether you are a $50k/month brand or a $5M/month brand — the math scales.
Important: do not include desktop. A mobile app does not cannibalize desktop sessions in any meaningful way. The cannibalization risk is between your mobile-web sessions and your mobile-app sessions, and even that is smaller than most founders fear. About 70% of app sessions are incremental — sessions that would not have happened without the home-screen icon.
Input 2: the install rate
How many of your mobile-web visitors will install your app? The honest answer depends on three things: how strongly you promote the app on your mobile site, how compelling the in-app offer is at install (first-order discount, exclusive launch, free shipping), and how loyal your audience already is.
For an unloyal cold-traffic brand, install rate sits around 1–2% of mobile sessions in the first three months. For a brand with a strong existing list and a push promotion at install, that rate climbs to 5–8% within six months. For brands with a cult following or a launch event behind the app, we have seen 10–15% in the first month alone, normalizing back to 5–7% as the easy wins are captured.
For modeling, use 3% as a starting point. If your brand has strong organic loyalty (cult followings, subscription bases, communities), use 5%. If you are mostly paid acquisition and your repeat rate is below 10%, use 1.5% and prove yourself wrong later.
Input 3: install-to-first-order conversion
Not every install becomes a buyer. Across the Shopify brands we work with, 60% of installs result in at least one purchase within 90 days. The remaining 40% either uninstall before purchasing or sit dormant. The 90-day window matters; many of your installs will purchase in week six or week eight, not week one. If you measure too early, you will underestimate the channel.
The single highest-leverage move on this number is the install-time offer. A first-order discount of 10–15%, gated on an app install and a single push opt-in, lifts install-to-first-order conversion from 60% to 75–80% in our data. Without an offer, expect 45–55%. Push opt-in matters because it is what lets you re-engage the 40% who lapse.
Input 4: repeat purchase rate among installed users
This is where apps actually pay back. Among users who install and purchase once, repeat purchase rate runs 30–50% higher than the same brand's mobile-web baseline. The mechanism is the install: home-screen presence keeps the brand top-of-mind, push prompts the return, and Apple Pay or Google Pay makes the second purchase a five-second action.
If your mobile-web repeat purchase rate is 22%, model app users at 30–32%. If your baseline is 35%, model app users at 45–50%. The lift is proportional; a brand with no existing repeat behavior does not gain magic repeat behavior just because it shipped an app.
“The install graph is an asset. It compounds. Every install you have today is a discount on every push campaign you will ever send.”— Appolar growth deck, 2026
Input 5: push notifications and incremental revenue
Push is the most under-credited revenue lever in mobile commerce. On a 100,000-install base, a single well-targeted campaign push routinely drives $20k–$50k in same-day revenue at zero per-message cost. Abandoned-cart push recovers an additional 8–12% of carts that would otherwise be lost. Replenishment push for refill categories adds another 10–15% to LTV.
For modeling, assume push contributes 15–20% additional revenue to your app channel above what the organic install graph alone would produce. That number sounds large; in practice it under-represents the contribution because push also boosts your repeat-purchase rate above baseline by reducing the friction-to-return.
Putting it together
Take a brand doing $2M annual mobile-web revenue, with a 22% mobile-web repeat rate, that ships a native app. Use the conservative inputs above: 3% install rate, 60% install-to-first-order, 32% repeat purchase on installs, 18% push uplift. Model the 12-month outcome: about 20–25% lift in total mobile revenue — roughly $400k–$500k of incremental annual revenue — plus a growing install base whose contribution compounds in year two.
Set against a build cost in the low five figures and a monthly platform subscription, the payback period typically runs 6–12 months for brands above $1M annual revenue and 3–6 months for brands above $5M. Below $500k annual revenue, the math is tighter; the app can still pay back, but it needs more aggressive install promotion to hit the volumes the model assumes.
What the model misses
Two things this model deliberately underplays. First, brand equity. A real app channel signals seriousness to your buyers, your investors, and your platform partners in a way that a mobile-web optimization cannot. That value is real and unmeasured.
Second, downside protection. Email deliverability gets worse every year, SMS regulation tightens every quarter, paid acquisition gets more expensive every cycle. The brands that own a direct push channel into their buyers' phones are insulated against the channels that compete for the same attention. Five years from now, that is going to look like the most important move you made.
Plug your own numbers into the model. If the 12-month payback math is solid, ship. If it is not, fix retention first, then come back. Appolar exists to make the ship part cheap; what we cannot do for you is decide whether the channel is worth your attention.
