Scale Million Dollar Shopify Apps Journey
10 years building GemPages from one person to 100,000+ stores and $1B/year in ad spend. How Shopify Apps are changing — and where the opportunity still is.
This past April I stood on a small stage in Shanghai, in front of about fifty people who were considering building Shopify Apps from China for global markets. The event was hosted by Cross-border Benny. The original slides I wrote in Chinese — the title was “打造百万美金 Shopify App 的’心法’” — the inner method for building a million-dollar Shopify App.
This post is a longer, more complete, more honest version of that 30-minute talk. I’m rewriting it for two groups: founders considering jumping into the Shopify ecosystem, and DTC operators who want to understand the funnel their ad dollars are actually flowing through.
The central question is simple: after 10 years, are there still opportunities in Shopify Apps — and if so, where?
Short answer: yes, but the game has changed. The longer version needs live data from the App Store, some expensive lessons I paid to learn, and the patterns repeating in apps that are winning in 2026.
A few numbers before the story
To give you a sense of GemPages’ scale today:
- 100,000+ active stores on Shopify
- 18,000+ brands paying us
- 1,000+ Shopify Plus stores — the high-end segment, very large revenues
- $1B+ in ad spend per year flowing through landing pages built with GemPages
The number I’m proudest of isn’t the customer count — it’s the commerce volume that flows through us. 18,000 paying brands, mostly DTC running heavy paid ads. A $100K/month brand burns more than half on ads, a Shopify Plus brand burns millions a month. All that traffic — Facebook, Google, TikTok — pours into landing pages built with GemPages. We’re not just a page design tool. We’re the funnel that billions of advertising dollars must pass through before becoming revenue.
GemCommerce — the parent company — has been bootstrapped for ten years. No funding rounds. Now 150+ people, headquartered in Hanoi, with team members in Singapore and across APAC. GemPages is the #2 Page Builder on the Shopify App Store. That’s the context for the rest of this piece — so it’s not theory.
10 years in one picture
Before going year by year, I want to show the full picture — especially for anyone starting out solo, because from 2014 through end of 2017, almost 4 years, I was completely alone.
| Year | People | Event |
|---|---|---|
| 2014 | 1 person | First idea |
| 2016 | Solo | First line of code |
| 2017 | Solo | First customer paid $99 lifetime |
| 2018 | ~5 people | Core team — first hire HR |
| 2019 | ~15 people | Moved into Cogo Coworking · raised price to $29 |
| 2020 | ~30 people | Covid · #1 reviews on Page Builder App Store |
| 2021 | ~50 people | ⚠ Security incident · 6 months banned · pivot to Freemium |
| 2022 | ~70 people | 💀 V7 crisis · zero growth all year |
| 2023 | ~80 people | Stabilize & rebuild momentum |
| 2024 | ~120 people | CRO direction · Sales Funnel |
| 2025 | 150+ people | V7.5 · 100K+ stores · ShopX 300+ |
| 2026 | 150+ people | GemAI Era · APAC 4 hubs |
The two pink-purple years are the hard period (2021 and 2022). The rest is waves up and down. No phase was a straight line. If I had quit in 2021 or 2022 — there would be no today.
2016–17 · One coder, one idea, the first $99
Successful products usually come from one of two groups: research the market then build, or solve your own problem. GemPages is the second group.
Before GemPages, I have to mention the 5 years before. From 2011 I was doing website customization — Drupal, Joomla, then WordPress. 5 years sitting with customers, hearing them complain that sites loaded slow, landing pages didn’t convert, ads were burning money without converting, and they wanted to change layout for a campaign but had to wait a week for a developer.
Those 5 years didn’t give me a pretty CV, but they gave me something more important: I deeply understood the pain points of online sellers. When I switched to Shopify in 2016 and couldn’t find a good-enough page builder, I didn’t need to research — I was that market for 5 years. The only question: “Why don’t I just build it myself?”
I coded GemPages completely alone. From late 2016 to nearly end of 2018 — only me coding, designing, supporting, talking to customers. No investor, no team. Just one belief: Shopify merchants need this tool.
Mid-2017 the first customer arrived — a $99 lifetime package, they bought immediately. That moment I realized: if a customer is willing to pay for a rough beta, the pain point is very real. I switched to subscription per Shopify guidelines. GemPages didn’t succeed because it was easy to use — it succeeded because it solved exactly the need of a very specific focus group: people like me, who knew code, had used WordPress page builders, and wanted the equivalent on Shopify.
The moment that kept me going
For nearly two years I literally lived at UPBK coworking — not a metaphor. Worked late, slept there, woke up and worked again. UPBK let entrepreneurs sleep over; for showers I went to a gym. Sounds insane, but at the time it was the most rational choice — saved time, saved money, stayed next to the product 24/7.
In those two years, the people I had hoped would join one by one walked away. There was a stretch where only I kept going. Revenue was still small, not enough to pay salaries and stay alive. There were months I didn’t know if I could keep going — the kind of difficulty solo founders all understand, that no one talks about, that only you face.
Then one day I got an email from a customer named Jie Kang, in the U.S.:
“I’m flying to Vietnam. I want to meet Chris.”
He wasn’t an investor or a partner. Just a user — paying $29/month — who loved the product. And yet he paid out of pocket, flew to a country he’d never been to, just to meet me.
Jie Kang didn’t help me with money — no investment, no big order. What he brought was something else, something I needed even more than money at that moment. When you’re alone and starting to doubt yourself, someone flying halfway around the Earth just to say “what you’re doing matters” — that’s what makes you get up the next morning.
Lesson: customers are not just numbers on a dashboard. If you’re genuinely kind to them, they will be the reason you don’t quit on the hardest nights. Years later I still tell this story to the team — to remind us why we do this work.
2018–19 · Forming the core team in coworking
2018 is when we formed our first core team. The interesting thing: first hire was not a developer — but HR. To build a company, I needed someone to handle people first; I could code myself, hiring people I couldn’t.
After HR, we hired interns from top universities in Hanoi — Antony (PTIT, dev), Nicolas (BK, full-stack), Janie (FTU, CS). End of 2019, we moved into Vincom Nguyen Chi Thanh, renting a corner in Cogo Coworking. We also raised price from $19 to $29 — really, this was filtering for serious users.
Out of this group, many became GemPages influencers. A story I remember: a French YouTuber named Frank made a GemPages intro video on his own — without being paid. Why? Because I personally supported him on every small question. In the same period, a competitor with more headcount started studying us, analyzing, learning — and shipped a similar product in a short time. Big pressure for a small team like ours.
2021 · Covid peak → 6 months banned
2021 is the biggest turning point in 10 years of GemPages — and the most expensive lesson.
Early 2021, Covid was at its peak, GemPages was growing at record pace — at that moment we had the most reviews of any Page Builder app on the Shopify App Store. Then an unexpected incident: GemPages had a “Mailchimp Newsletter” element that didn’t comply with a Shopify security regulation we didn’t know about. Result: Shopify banned GemPages for 6 months — no new installs.
Those six months were brutal: product frozen, team morale crashed, competitors took the entire share. There was even a Facebook Group trading Shopify accounts that had GemPages already installed — which also showed how much GemPages had been loved.
When we came back end of 2021, the market had shifted. Competitors had more customers → faster bug fixes → more polished product → more customers. A loop we couldn’t catch up to. Research showed most Shopify Apps had moved to Freemium. Decision: free the first page. A required move to win back the game.
Two lessons from 2021. One, if you build for global markets — especially the West — you must understand law, security, compliance from day one. Your product can be great, but a small slip can push you out of the game. Two, as a small startup, you almost have to put all eggs in one basket. When we were banned for 6 months, we didn’t pivot, didn’t jump to a different platform. We stayed on Shopify and fixed what needed fixing — because that was the only basket. Diversification is for those who already won. When you haven’t, focus is the only thing that saves you.
2022 · The V7 crisis — “the bloody road”
After the 2021 hit, we thought simply: “We have to rebuild stronger.” V6 used old JavaScript, didn’t scale, the editor lagged. Decision: rewrite from scratch. GemPages V7. Looking back, technically right — and a bloody road on everything else.
The day V7 went live, within hours, complaints poured in like hailstones. The rewritten code still had many bugs. UX was different from V6 — old customers opened it and didn’t recognize the familiar tool. Agencies sent long angry emails because V7 slowed their delivery. While firefighting V7, V6 stood still — we had thrown 100% of resources at V7, no one was maintaining V6. Competitors filled that vacuum fast.
There were stretches in 2022 where I’d come into the office, sit reading tickets, and not know which to fix first. Everything was on fire. The most painful wasn’t the bugs — it was reading feedback from customers who used to love GemPages, now writing that they were disappointed. To build a product for six years to be loved, then break it with your own hands — the feeling is hard to describe.
But the hardest wasn’t the product. It was the team. I have to admit the bigger mistake: at that moment I thought “the company is bigger now, we need stronger people, more senior people.” We hired a bunch of “stars” — pretty CVs, big company names, senior titles. But the opposite happened: they didn’t help much, they made the system messier. Old people couldn’t work with new people. Conflict escalated faster than any V7 bug.
Result: 2021–2022 many key team members left. Some had been there since the early Cogo days. Some mornings I’d come into the office, see another empty desk, ask myself: “Am I going to end up alone again like 2016?”
A better product doesn’t mean customers will accept it.
The most expensive lesson from V7 isn’t “don’t rewrite,” but two things: a better product isn’t necessarily welcomed — customers have a learning curve, workflows they don’t want to abandon. And hiring “stars” is not the answer — if they can’t gel with the people who walked the journey from the start, you lose both.
If you’re in a year where nothing goes up, users don’t grow, the team contracts, and you’re doubting your biggest decision — that’s normal. Startups don’t always go up. There will be 1–2 years of no growth. That’s when persistence matters most.
2024–25 · Revival — pick one thing to do best
After hitting bottom on V7, we had to sit down and be honest with ourselves. GemPages had no clear differentiation in the market: a competitor had more users, a competitor targeted high-end, a competitor was the cheapest. GemPages? We belonged to none. Hardest lesson from V7: trying to do everything is the fastest way to be good at nothing.
The only question: “If we could only pick ONE thing to do best, what does GemPages pick?”
Answer: conversion. Repositioned — “#1 Conversion-focused Page Builder for Shopify”. Every product decision after that ran through one filter: “does this help raise conversion?” If not, we don’t build it. We shipped a series of deep CRO features — Sales Funnel, Post-purchase Upsell, Headless Landing Page, Image-to-Layout AI.
In parallel, 2025 we launched V7.5 — inheriting the “soul” of V6 + the power of V7. The first version in four years we could confidently say was “better than V6.” Growth came back. Lesson: when you can’t win by being “bigger,” win by being “more different”. Find the point where you can be best, and pour your heart into it.
6 inner methods after 10 years
This is the part the Shanghai room asked the most about. The six lessons I hold dearest after 10 years of GemPages.
1. Ikigai — find the core value. The intersection of four things: what you’re good at, what you love, what the world needs, what people will pay for. For me: 5 years of ecom + good at code + know design → page builder. Stand at the intersection where you’re strongest. Don’t build a product just because of a trend.
2. Do one thing best. At first do only one thing well: page builder. Don’t think about funnels, AI, or ecosystems yet. When solid, expand. Don’t do 10 things at once — pick 1 and do it better than anyone.
3. Customer service decides the game. Tony Hsieh, Zappos founder: “Customer service should not be a department, it should be the entire company.” In the first two years, I stayed up till 2am every night taking care of customers. Some days I’d leave the office and see the sunrise — only then did I notice that life had its beauty too. Many times a customer changed the future of the startup just because of one kind conversation.
4. Stand on the shoulders of giants. A wrong belief I often hear: “Building on someone else’s platform is risky, you need your own platform.” I don’t think so. Building on Shopify isn’t bad — leverage the platform’s advantage to go global. For small startups: put eggs in one basket and do it well. Diversification is for the already-grown.
5. Move fast, but with momentum. Every startup feels the pressure to move fast. But fast must come with system and stability. Build process first, scale headcount second. Lesson from us: V7 crisis, we tried to hire more to “go faster” — broke our momentum.
6. Trust your team. No one is complete — that’s exactly why we need a team. Internal lack of trust → very hard to go far. Every time someone left over conflict, I asked myself: “Did I trust them enough?”
Build or Buy team?
A question I get often. My view is clear — for small startups: build, don’t buy. “Stars” are usually expensive, carry high expectations and old playbooks, and easily clash with the team you have. They don’t necessarily understand the product or the customer — big-company experience doesn’t transfer 1-to-1 to a startup. Startups need spirit and drive, not skill. Skills can be taught, drive cannot.
My first hire was HR, not a developer. After that, all interns from top Hanoi universities. They had no experience but had huge drive to grow. After 5–7 years, many became leaders, even CTOs. Process first, people second — otherwise you go in circles. That’s the quietest startup killer.
CRO Evolution 2015 → 2026+ — 5 eras
Before the App Store data, I want to pause on a key concept for anyone in ecommerce: CRO isn’t small website tweaks — it’s strategy for survival. In 10 years, CRO has moved through 5 waves.
| Era | Years | Characteristics |
|---|---|---|
| 1 | 2015–17 | Default themes. CRO = trust badges, checkout speed. “Having an online shop” was enough. |
| 2 | 2017–19 | Custom Landing Pages. Dropshipping explodes. CRO shifts from “website” to “funnel”. |
| 3 | 2020–22 | Standard funnel + UI/UX. Upsell, bundle, cross-sell, subscription. Storytelling rises. |
| 4 | 2022–24 | Advertorial & Pre-sell. Ad traffic expensive. Brands warm up before selling. |
| 5 | 2026+ | AI & Personalization. CRO must also “talk” to AI shopping agents. |
The essence of CRO doesn’t change — still helping the customer overcome the barrier to a decision. But that barrier shifts every year. 2015 asked “Is this site trustworthy?” → 2024 asks “Why this brand and not the competitor with the same offer?” → future: “Will the AI agent trust this brand enough to recommend it?”
Practical consequence: don’t push cold traffic straight to the sales page. The 2026 customer needs to warm up first — understand the problem, trust the story, then look at the offer. Top brands now run ads to pre-sales pages (advertorial, listicle, native story) before sending traffic to a sales page.
5 highest-converting advertorial archetypes
Our team analyzed 1,581 advertorials currently live on high-revenue Shopify stores. Common trait: they don’t look like ads. Five main formats at top of funnel:
- Listicle “5 Reasons” — tell the problem first, the product reveal at the last reason. Converts 3x a traditional page.
- Advertorial “The truth about…” — medical-style article, educate first, reveal product as a “discovery.” Strong in health/skincare. The dominant format for 2026 cold traffic.
- Native Story “How I…” — looks like a news-site article. Cold traffic doesn’t realize they’re seeing an ad.
- Comparison “Top 10” — serves comparison intent. Your brand sits at #1.
- Case Study “How Sarah fixed it” — customer story, before-after, dialogue. Strong in health/beauty/weight-loss.
Those 5 advertorials only do one job: get the click. After that click, if the sales page is weak — ad money burns to ash. The 5 best closing sales-page layouts: Long-form (hero → benefits → proof → FAQ → offer); PDP style (gallery + variant + add-to-cart); Single-hero offer (bundle + urgency); Story-to-sale (long story → product as solution); Benefit-stack (numbered benefits, easy to scan, mobile-friendly). No format wins universally — test 2-3 layouts in parallel, scale the winner.
Shopify App Store today — live data
The answer to “Are there still opportunities in Shopify Apps?” shouldn’t rely on intuition. I want to answer with live data from the Shopify App Store, crawled directly from 2,211 apps and 161 categories a few hours before the Shanghai event. Sources: apps.shopify.com sitemap + Shopify Investor Relations Q4-2025 + Marketplace Pulse.
The App Store today in 4 numbers
- 17,629 apps in the sitemap
- 161 categories
- $1B/yr Shopify pays developers (2024) — cumulative now over $1.5B
- 2M+ active merchants, 47,000 Shopify Plus with ARPU 5–10x SMB
Power law — extreme
The Shopify App Store is winner-take-most:
| Tier | # of apps | % of reviews |
|---|---|---|
| Top 1% | 22 apps | 24.3% |
| Top 5% | 110 apps | 52.8% |
| Top 10% | 220 apps | 68.4% |
| Bottom 75% | 1,586 apps | 12.5% |
| Freemium model | — | 46% (paid-only just 1.7%) |
The 220 apps in the top 10% eat 68% of the pie. The other 17,400 apps share 32%. Not to discourage you — but to make clear: you can’t “build a generic app” and win. You have to pick one niche. And launching paid-only today = losing half the funnel. Freemium is a survival requirement.
Top 10 store-wide + Page Builder Battle
The #1 app on the entire Shopify App Store today is Judge.me (37,406 reviews, ⭐5.0) — a product reviews app from Hong Kong. Remember this detail: the #1 App Store app comes from Asia — not Silicon Valley, not New York. Same story repeats in China (DSers, 17TRACK), Vietnam (PageFly, GemPages, EComposer, Avada), Israel (Yotpo, Loox).
| Rank | App | Reviews |
|---|---|---|
| #1 | Judge.me — Reviews 🇭🇰 | 37,406 |
| #2 | TikTok — Channel | 12,816 |
| #3 | Shopify Flow | 9,270 |
| #4 | Pop Convert | 8,490 |
| #5 | Loox — Reviews + AI | 7,571 |
| #6 | SendWILL — Email popup | 7,153 |
| #7 | Appstle — Subscriptions | 6,498 |
| #8 | Shop App | 6,336 |
| #9 | PageFly — Page builder | 5,739 |
| #10 | DSers — Dropshipping 🇨🇳 | 5,377 |
Page Builder Battle — GemPages’ home turf: PageFly leads with 5,739 reviews; GemPages #2 with 3,601; EComposer #3 with 3,435 (only 166 reviews behind GemPages — the race isn’t over).
Lesson: don’t try to take #1. Don’t fight Klaviyo, don’t fight Judge.me. Aim for #3 to #5 in a niche you understand most deeply. Top 50 Shopify apps all have ARR >$10M. You don’t need to be Klaviyo — you just need to be Klaviyo for a specific niche.
Three opportunity zones — money slide
Analyzing 2,211 live apps surfaces three strategic zones for anyone wondering which category to pick:
Zone 1 — Quality gaps. Categories with low rating + high reviews = users suffering because there’s no better choice. Top 4: Operations ⭐4.38 (24 apps · 21K reviews), Selling Online ⭐4.39 (24 apps · 42K reviews), Inventory ⭐4.41 (24 apps · 10K reviews), Sourcing ⭐4.48 (24 apps · 23K reviews). The store-wide average rating is 4.72 — these run 0.3-0.4 stars below. Build an app with 20% better UX → take share in 6-12 months.
Zone 2 — Rich niches. Few apps, high concentration of reviews — sweet spot. Social Trust (18 apps × 3,423 reviews/app), Email Marketing (12 × 3,104), Pre-order (15 × 1,137), Print on Demand (17 × 1,122). Low supply, high demand — newcomers can climb to #3 in a few years.
Zone 3 — Low-barrier entry (low BFS). “Built for Shopify” badge is a high standard, averaging 43%. But Operations is only 29% BFS, Finance 33%, Orders & Shipping 38%. Just hitting BFS already puts you in the top 30%.
Golden slot — intersection of all 3 zones: “Operations apps”. Lowest rating (⭐4.38), lowest BFS (29%), 21K reviews (demand proven). Largest gap in the data. Specifically: order-management automation, bulk operations, workflow — Chinese sellers are doing this by hand and suffering.
Money heatmap — which segments are HOT, which are COLD
Tiers based on avg reviews/app — the strongest proxy for install velocity and revenue:
| Segment | Apps | Avg/app | Rating | BFS | Tier |
|---|---|---|---|---|---|
| Marketing & Conversion | 37 | 2,894 | 4.65 | 49% | 🔥 HOT |
| Store Design 💎 | 30 | 2,978 | 4.64 | 57% | 🔥 HOT |
| Notification | 24 | 1,883 | 4.85 | 75% | ✓ GOOD |
| Sales Channel | 35 | 1,584 | 4.70 | 77% | ✓ GOOD |
| Selling Products | 37 | 1,530 | 4.71 | 86% | ✓ GOOD |
| Operations ⚡ | 24 | 906 | 4.38 | 29% | ⚡ GAP |
| Storefront | 24 | 900 | 4.62 | 58% | ~ OK |
| Finding Products | 27 | 844 | 4.74 | 63% | ~ OK |
| Internationalization | 25 | 723 | 4.80 | 52% | ~ OK |
| Payment | 31 | 702 | 4.70 | 65% | ~ OK |
| Orders & Shipping ⚠ | 39 | 622 | 4.52 | 38% | ⚡ GAP |
| Support | 33 | 595 | 4.79 | 61% | ~ OK |
| Store Management | 39 | 588 | 4.74 | 54% | ~ OK |
| Finance | 24 | 543 | 4.65 | 33% | ⚡ GAP |
| Security | 27 | 442 | 4.86 | 63% | ❄ COLD |
| Content | 24 | 316 | 4.75 | 67% | ❄ COLD |
🔥 HOT = prints money but crowded. ✓ GOOD = still room for newcomers. ⚡ GAP = where cross-border founders should stand (low rating + low BFS = clear quality gap). ~ OK = nothing notable, skip unless you have domain expertise. ❄ COLD = don’t enter unless you have a specific moat.
Shopify Plus — the 47,000-enterprise gold mine
The opportunity Asian developers pay the least attention to. Shopify Plus has 47,000 merchants globally: 41K US (85%), 6K Canada, 5.5K UK, 3K Australia, 2.5K Germany. Singapore + APAC growing but under-served.
Why does Plus matter? Plus ARPU is 5–10x SMB. An SMB store pays $30/month, a Plus store pays $300–3000/month for the same app. 1 Plus customer ≈ 50 SMB customers. And competition is lighter — only ~3.9% of apps have the “Works with Shopify Plus” badge, roughly 500–700 apps out of 17K.
Five concrete Plus opportunities for cross-border devs:
- B2B / Wholesale — 2025 Shopify B2B GMV grew +96%. TAM ~$50M ARR. Competitors: BSS B2B, SparkLayer — clearly room left.
- Checkout Extensions — Shopify is pushing Checkout Extensions over legacy. Old apps haven’t ported in time — a 1-2 year window of gold.
- Multi-warehouse / Multi-region — sweet spot for Chinese devs. Plus ships globally, needs to sync 1688 + Shopify + 3PL. No one understands 1688 better than Chinese devs — moat built-in.
- ERP integrations — manufacturing Plus brands need to connect to SAP, NetSuite, Infor M3. Apps are seriously lacking. Build a good connector → charge $500–2000/month.
- Enterprise security & compliance — GDPR, SOC2, fraud detection for high-volume Plus. SMBs don’t pay much for security — Plus pays a lot.
SMB market = volume. Plus market = margin. 1 Plus customer = the value of 50 SMB customers — and competition is 5x lighter.
The brutal cliff — and the death zone
Two numbers will change how you see the App Store.
The cliff. Real drop-off curve — ranking 2,211 apps by review count: #1 Judge.me 37,406; #10 DSers 5,377 (down 86%); #100 Rivo Loyalty 1,368 (3.7%); #1,000 C:Hub 71 reviews (0.2%); #2,000 NFTeapot 3 reviews (near zero). Rule: every 10x in rank, reviews drop ~90-95%. There’s no shortcut to the top.
Death zone. In a sample of 2,211 apps (which is itself the top tier): 75 apps have 0 reviews (3.4%, “zombies”), 401 apps have 1-9 reviews (18.1%, “struggling”). Total 476 / 2,211 = 21.5% are dead or dying. In the full 17,629 apps, estimate 40-50% are zombies.
The real market isn’t 17,629 apps — it’s only ~5,000–7,000 apps that someone is actually using. If you make top 1,000 (just 71 reviews), you’ve already beaten 90% of the real market.
5 patterns in every winning app
Not theory — patterns that repeat in the data:
1. Asia is WINNING. Top apps come overwhelmingly from Asia: Judge.me (HK) #1 store-wide with 37K reviews, 17TRACK and DSers (China), PageFly + GemPages + EComposer + Avada (Vietnam), Loox + Yotpo (Israel). You don’t need to be in Silicon Valley to build the #1 app.
2. Specialists beat generalists. Top apps each do one thing: Judge.me only reviews; Appstle only subscriptions; Kaching only bundles; STOQ only pre-order; Pandectes only GDPR. No top app is “all-in-one”. Pick 1 thing, do it best in the world.
3. AI wave — 13% of apps already renamed. 288/2,211 apps now have “AI” in their name. 5 years ago this number was near 0. Top: Loox AI 7.5K, Avada AI 4.4K, BOOSTER AI 4.3K. Not too late.
4. 3rd-party beats Shopify Native. Pop Convert (8,490) > Shop App (6,336). Loox (7,571) > Shopify Inbox (5,302). Judge.me (37,406) >> all Native combined. Shopify can’t build everything itself. You still have a window.
5. Freemium = 46% · Paid-only is dead (1.7%). Sellers in 2026 don’t trust apps without a free tier. Launching paid-only = self-extinction.
These 5 patterns — Asia + Specialists + AI + 3rd-party beats Native + Freemium — are the recipe for winners. They line up 100% with what GemPages has done over the past 7 years.
AI = Software 2.0 — a new opportunity window
Before AI, Page Builders were genuinely hard for normal people. Even at GemPages, only ~5% of users could actually design a great-looking page. AI changes that. The new message: “Anyone can design a landing page and optimize for conversion.” GemPages led with Image-to-Layout — upload a screenshot, AI builds an editable page. Our intro video from 2 years ago has nearly 1 million views, and hundreds of thousands of sections have been generated by AI.
But AI is also a massive challenge. If we don’t innovate, GemPages will be replaced. That’s why we’re going all-in on GemAI — CRO Brain: a foundational AI layer that connects GemPages, GemX, and Gemians into a self-learning, self-optimizing CRO brain for each store.
Three final arguments for “Are there opportunities in Shopify Apps?”:
D. A 1–5 person team can beat a team of 100. In the next 1–2 years, most coding will be done with AI assistance. What used to need a team of 5–10 can now be built by 1 person + AI in a few weeks. The barrier and cost of dev are at their lowest in history.
E. The advantage goes to “real builders”. The winners are builders who deeply understand cross-border — not people who just know how to use tools. If you’re in China you already have the edge: you understand Chinese sellers, real cross-border pain points. This is a moat Western founders don’t have.
F. App → AI Agent is the next wave. Shopify Apps are shifting from “tools” the user clicks to “agents” that work on the user’s behalf. Customers will keep getting lazier about clicking — they just want to say “do this thing” and have an agent finish it. Catching this wave = a very large business. The next 1–2 years matter.
The Shopify App Store is at the end of Growth, headed into Maturity. The opportunity window is still open — but every major disruption (like AI) can reset the curve, creating a new wave. This is a special moment.
If you’re building a Shopify App, ask yourself: “If an AI Agent can do my job, why do I still exist?” Build that future before it builds you.
Great product + excellent CS is no longer enough
Early on, the SaaS formula was simple: great product + great CS → growth. That’s how we survived to 2021. But 2026 — the context is completely different. Every good app does that. The App Store is crowded, organic traffic is expensive, Shopify controls distribution. If you only compete on ranking and reviews, you’ll burn out.
To create a durable advantage, you need a distribution layer of your own — channels, community, moats that don’t sit on the Shopify App Store. That’s why we organize ShopX — “The Ultimate Gathering for E-commerce Professionals”. ShopX is not marketing — it’s infrastructure.
Current scale: 300+ attendees/year, 50+ Shopify Plus brand operators, 100+ app partners, 4 APAC cities. 2026 will scale to 500 and expand to Singapore and Shanghai. Benefits: direct relationships with high-tier customers; network effect; brand authority — people associate “GemPages” with “CRO leadership.”
Advice: early on, think about your distribution layer. Don’t just build the product — build the community around the product. That’s the real moat in 2026.
So — are there opportunities in Shopify Apps?
Yes. But the game has changed. Six points distilled from 10 years inside + live App Store data:
A — Crowded, but most apps are weak. Alongside a few that are very good, there’s still a lot of room for a more polished product to slip in.
B — From local → global. Serve local first (not just Chinese — also Japanese, Korean, Spanish, Brazilian, Indonesian, Vietnamese). That’s a moat US/EU can’t touch.
C — Pick 1 segment, do it best. Don’t try to serve everyone. Pick 1 niche, understand the user, pour your heart into it. Different and deeper, not bigger.
D — AI opens the door for very small teams. 1–5 people + AI can beat a team of 100. Dev barriers and costs are at their lowest in history.
E — Advantage = a builder who deeply understands cross-border. You don’t lack opportunity — you only need a bit of courage to start.
F — App → AI Agent is the next wave. Catching it is a big opportunity. The next 1–2 years matter.
Two final lines
Before stopping, the two lines I want to leave you with.
One — focus, focus, and focus. Focus is daring to drop in order to chase the core. Accepting losses on a few fronts to win on the one you’ve chosen. It’s very easy to lose focus to the next “big thing.” Always remind yourself to focus on product, customer, and your strength.
Two — it’s not faster who wins, it’s the one who lasts. A startup is a marathon, not a sprint. Ten years of GemPages taught me that. There were 1–2 year stretches with no growth, but if I’d quit, there’d be no today. What matters is staying steady, stepping forward little by little. 1% better every day — after 1000 days you’re a completely different person.
My story isn’t perfect — there were moments near bankruptcy, years of zero growth, decisions I regret. But after 10 years, the biggest lesson: if you focus deeply enough, last long enough, are sincere enough — the result will come.
Hope the story is fuel for anyone in the early stretch of a similar journey.
This piece is the longer version of a 30-minute sharing at the Cross-border Benny event in Shanghai, 2026-04-11. If you’re building a Shopify App or a cross-border DTC brand and want to connect — LinkedIn or chris@gemcommerce.com.
Shopify App Store data: crawled directly from apps.shopify.com sitemap (April 2026, 2,211 apps × 161 categories) + Shopify Investor Relations Q4-2025 + Marketplace Pulse. Sample is top tier — full-store power law may be even more extreme.