A Cautionary Tale on Platform Dependency: Why Our Shopify App Went From Daily Installs to Zero Overnight

Building on someone else's platform feels like a smart shortcut. Shopify, Umbraco, WordPress, Salesforce, whatever you choose, you get instant access to an established marketplace, built-in payment processing, and users actively looking for solutions. No need to build authentication, hosting infrastructure, or payment systems from scratch.

Brilliant. Until it isn't.

I want to talk about the double-edged sword of platform dependency, what happened when installs for our Shopify reporting platform TrendSeam suddenly stopped, and why the Apollo story should keep you up at night if you're building platform-dependent products.

The Six-Week Currency Window

When we acquired Igloo (our Umbraco site builder package), we noticed something that changed how we think about platform products.

If you don't have a package supporting the most recent version of Umbraco within six weeks of release, you see a significant drop in installs.

Six weeks.

Fast-forward six months, and you're dealing with a handful of installs at most. Even if you're supporting a long-term support version, the organic growth stops.

This isn't unique to Umbraco. We've seen the same pattern with Shopify, and suspect it's pretty much the same for every platform with an active marketplace. Users want current versions. Platforms promote current versions. Fall behind, and you disappear.

Building a platform-dependent product isn't a one-time development effort. It's an ongoing commitment to stay current or face obsolescence. Factor that into your ROI calculations from day one.

The TrendSeam Story: From Daily Installs to Silence

We launched TrendSeam for Shopify in 2017. Back then, the app marketplace was much smaller. We were one of relatively few analytics apps.

Multiple daily installs. Minimal marketing effort. The marketplace was doing the work for us.

Fantastic feeling.

Then we got a couple of reviews mentioning that the API integration was breaking. We'd fallen behind on updates. Nothing catastrophic, just not keeping pace with Shopify's API changes.

The installs stopped as quickly as they'd started. Not gradually. Not a slow decline. Just stopped.

Shopify's algorithm noticed the negative reviews and the outdated API usage. We dropped in marketplace rankings. New merchants never saw us. The daily growth evaporated.

We've recently released an updated version and are now discussing whether releasing new versions is the way to handle this, or whether we need a different approach entirely.

The lesson? Platform currency isn't optional. It's existential.

The Apollo Cautionary Tale

If you're not familiar with Apollo, it was the go-to mobile app for browsing Reddit until mid-2023. Nicely designed. Hugely popular. Speculated to be a multi-million pound business.

Reddit decided to monetise API access. Fair enough, it's their platform. But they priced it in a way that made Apollo's operation completely untenable.

Overnight, Apollo went from thriving business to dead.

Not because the product was bad. Not because users stopped wanting it. Because the platform changed the rules.

Christian Selig (Apollo's developer) shut it down rather than operate at a loss. Years of work. Millions of users. Gone because Reddit's management priorities shifted.

This is the nightmare scenario for anyone building on someone else's platform. You don't control the economics. You don't control the API. You don't control your destiny. One policy change, one pricing adjustment, one strategic shift at the platform level, and your business can disappear.

When Platforms Turn Against You

You don't need a dramatic API pricing change to kill your product. Sometimes platforms simply shift what they promote.

We've watched successful apps fall out of favour when platforms launched competing features. The platform's own solution gets promoted. Third-party apps get buried in search results.

Or the platform changes its review algorithm, and negative reviews suddenly carry more weight. Or they adjust marketplace categories, and your app ends up in a less-trafficked section.

You're dependent not just on the platform continuing to exist, but on them continuing to support your position in their ecosystem. That's a lot of dependencies for something you're betting your business on.

The Benefits Still Outweigh the Risks (Sometimes)

I'm not arguing against building platform-dependent products. We're still doing it. TrendSeam is Shopify-dependent. Igloo is Umbraco-dependent.

The benefits are real. Faster time to market because you skip building infrastructure and focus on your core functionality. Built-in distribution through marketplaces with active users already searching for solutions. Lower customer acquisition costs because the platform does marketing work for you. Established trust, since users trust apps in official marketplaces more than random websites. Payment infrastructure without handling billing, subscriptions, or payment processing yourself.

For an agency exploring products, these advantages can mean the difference between shipping something and abandoning it halfway through.

But go in with eyes open about the risks.

How We Manage Platform Risk

Here's what we do at TSD to reduce (not eliminate) platform dependency risk.

First, we budget for ongoing updates. When we calculate product ROI, we assume spending 20-30% of initial development time annually just staying current. New platform versions, API changes, security updates. This isn't profit. It's maintenance. Budget for it.

Second, we monitor platform health. If your chosen platform is declining, your product is too. We watch platform announcement blogs, developer communities, and marketplace trends. Umbraco releasing more frequent versions? We need to speed up our update cycle. Shopify launching competing features? We need to differentiate or pivot.

Third, we build platform-agnostic where possible. TrendSeam's analytics engine isn't Shopify-specific. The data processing, segmentation logic, and reporting could work on other platforms. The integration layer is Shopify-dependent. The core value isn't. If we ever needed to port TrendSeam to another platform, we wouldn't start from scratch.

Fourth, we diversify revenue. Don't bet your entire agency on one platform-dependent product. We run multiple products across different platforms. If Shopify changes something that kills TrendSeam, we've still got Igloo on Umbraco. If Umbraco shifts direction, we've got other revenue streams. Diversification doesn't eliminate risk, but it stops any single platform change from destroying your business.

Fifth, and this is the big one, we maintain agency revenue. We've never tried to replace agency income entirely with product income. Products supplement agency revenue. They bridge quiet months. They provide alternative income streams. But they're not the whole business. If every product failed tomorrow, TSD would still operate. That breathing room lets us take risks we couldn't take otherwise.

When Platform Dependency Makes Sense

Despite the risks, here's when building on platforms is worth it.

You're an agency exploring products, and the reduced complexity and faster time to market matter more than long-term platform risk. The platform is growing, and a rising tide lifts all boats. Your differentiation is strong, so you're solving a problem the platform won't or can't address, making you less vulnerable to them launching competing features. You've got ongoing development capacity, so staying current is manageable because you've already got developers on the team. Or you're treating it as learning, since first products don't need to last forever. Getting experience shipping and maintaining a product is valuable even if it eventually gets shelved.

When to Avoid Platform Dependency

Walk away if the platform is declining. Building on a shrinking platform is fighting gravity.

Walk away if you can't commit to staying current. If keeping pace with updates isn't realistic, don't start. You'll just watch it die slowly.

Walk away if your core value duplicates platform features. If the platform could easily build what you're building, they probably will.

Walk away if you need certainty. If losing the product would sink your business, don't make it platform-dependent.

What We're Doing Differently Now

The TrendSeam experience taught us to approach platform products differently.

We now have a "platform health scorecard" we review quarterly. Simple spreadsheet tracking platform release frequency, marketplace growth rate, competing features launched, our app review trends, and install and churn rates.

If metrics start trending wrong, we decide: double down with updates, pivot the features, or start planning an exit.

It's not sophisticated. But it forces us to pay attention before problems become catastrophic.

We're also more selective about which platforms we build on. Shopify and Umbraco feel relatively safe, large communities, active development, clear roadmaps. A smaller platform with uncertain futures? Probably not worth the risk.

The Bottom Line

Platform dependency is a calculated risk, not a guaranteed disaster.

You get faster time to market and lower initial costs. You accept ongoing maintenance demands and existential platform risk.

For agencies exploring products, this trade-off often makes sense. You've got development capacity. You can stay current. The reduced complexity means you're more likely to actually ship something.

Just don't fool yourself into thinking you're building an independent business. You're building on someone else's land. They own the ground beneath you.

Budget accordingly. Stay current religiously. Watch for platform shifts. And maintain revenue streams that don't depend on any single platform's goodwill.

Have you built platform-dependent products? What happened when the platform changed something major? I'd like to hear the war stories.