Inside LeapOut’s Hardest Shopify Plus Builds: Regulated Commerce, Enterprise Scale, and Going Headless By Marvin Ortiz, Co-Founder and Managing Partner, LeapOut Digital  The short version: Most agencies show you their prettiest work. We’d rather be judged by our hardest. This is a deep look at three Shopify and Shopify Plus builds that each solved one of the three hardest problems in enterprise ecommerce — selling a regulated product (MaxiLife by Maxicare), executing a global brand’s standards at scale (Under Armour Philippines), and extending Shopify Plus past its native limits into a headless build (Kotis Design, USA). We chose these three because difficulty is the one thing a portfolio can’t fake. If you want to know whether a team can actually build, look at what it does when the easy path runs out.   A portfolio full of beautiful storefronts proves almost nothing. Anyone with a good designer can produce a clean store on a forgiving brief. What separates a real engineering partner from a theme shop is what happens when the brief isn’t forgiving — when a regulator is involved, when a global brand sets a standard you can’t bend, or when the platform itself says “no.”  So instead of showing you everything we’ve built, I want to go deep on three. Not our prettiest work — our hardest. Each one represents a different way an ecommerce build can become genuinely difficult, and together they’re the closest thing we have to an honest answer to the question every serious client is really asking: can you handle the part that’s hard?  Here they are. Build One — MaxiLife by Maxicare: Selling a Regulated Product Online The problem most agencies won’t take. MaxiLife by Maxicare is a regulated, healthcare-adjacent insurance product from one of the country’s largest health-maintenance organizations — sold, for the first time, through ecommerce. That single fact changes everything about the build.  Why it was hard. Selling a regulated financial-and-health product isn’t like selling apparel. The build has to satisfy disclosure, compliance, and documentation requirements that a normal store never encounters — and it has to do that without turning the purchase into a punishing legal form. The entire challenge is a contradiction: make something heavily regulated feel light and human to the person buying it. Get the compliance wrong and you can’t launch. Get the experience wrong and no one buys. You have to win both.  What we did. We extended Shopify Plus with deep technical customization to meet the regulatory requirements while protecting the buying experience — building the compliance into the platform rather than bolting it on top, so the rules were satisfied structurally instead of being patched in. Precision wasn’t a preference here; it was the entire job.  What it proves. When we tell a prospect “we handle regulated commerce,” this is the build we point to — and it’s why brands in insurance, health, and finance take our calls. Regulated ecommerce is a specialist capability most agencies quietly avoid, and the avoidance is the opportunity.  “Your professionalism, dedication, and excellent service have been greatly appreciated… It’s been a pleasure collaborating with your team, and I truly value the strong relationship we’ve built. I will certainly recommend your services moving forward.” — Carlo Rodelas, MaxiLife, Digital Channels Manager Build Two — Under Armour Philippines: Executing a Global Standard, Flawlessly The problem you don’t hear discussed. Under Armour Philippines was one of the most demanding Shopify environments we’ve handled — and the difficulty was a specific, underrated kind: building to a standard we didn’t set.  Why it was hard. When you work with a global brand, the design language, the brand controls, and the performance expectations are all defined elsewhere, and they are non-negotiable. Your job isn’t to invent — it’s to execute someone else’s standard, locally, at the exact quality they require, every single time, while making the catalog, pricing, and promotional logic work for the Philippine market. A lot of agencies are good at being creative. Far fewer are good at being faithful — at delivering precisely what a global brand demands without drift or compromise. Shopify Plus gave us the flexibility; the scale demanded governance, because flexibility without structure creates risk at exactly the moment a global brand is watching.  What we did. Deep front-end customization aligned to global brand standards, disciplined performance engineering, and careful stakeholder alignment across local and global teams — the unglamorous governance work that keeps a high-traffic enterprise store fast, on-brand, and predictable.  What it proves. Being trusted by a global brand to touch its storefront is a credential in itself. Global and enterprise brands run procurement, legal, brand-safety, and performance reviews most local businesses never will. Clearing that bar is harder than winning any award — and it’s a bar we’ve cleared repeatedly. Based on Page speed Insights Report from Jun 14, 2026, 7:18:31 PM  Build-quality scorecard (Google Lighthouse): SEO 100 · Accessibility 95 · Best Practices 92 · Performance 79. A perfect SEO score and near-perfect accessibility are the marks of a build engineered to be found and usable, not just to look good — exactly the disciplined, measurable execution a global brand requires. Build Three — Kotis Design (USA): When the Platform Says No The problem at the technical frontier. Kotis Design is a US-based B2B company — a PPAI 100 firm, one of the largest distributors in the American promotional-products industry — serving major corporate clients with bespoke swag and merchandise programs. Their requirements exceeded what Shopify does natively. The platform, in effect, said no.  Why it was hard. Kotis needed heavy, per-client customization — bespoke corporate stores, redemption sites, and ordering flows tailored to each enterprise client. Shopify’s standard theme-and-app architecture doesn’t bend that far. A weaker partner says “Shopify can’t do that.” We treated it as the brief.  What we did. We built custom functionality to support complex product personalization, and as Kotis’s ambition for their platform grew, the work evolved toward a headless architecture — decoupling the storefront from Shopify’s native layer to deliver experiences and client-specific functionality the standard stack can’t, while keeping Shopify as the commerce engine underneath. It’s not a finished project; it’s a living platform we build against in regular sprints, and have for two years.  What it proves. Two things, and both

A few weeks ago I watched a query I’d normally pay good money to win get answered before my ad ever had a chance to compete. A health-conscious buyer typed a long, specific question into Google. No ten blue links. No tidy row of shopping ads up top. A paragraph. Two products named. A recommendation, formed and half-decided, before a single click left the page. That moment is the whole story of Google Ads in 2026. For fifteen years, optimization meant tuning the knobs on a machine you could see: bids, match types, ad copy, landing pages. You worked the auction and the auction worked back. That game still exists. But it’s no longer where the leverage is. Here’s the thesis I’d stake the year on: tuning your bids and keywords is no longer how you win Google Ads — it’s just the price of competing. The real leverage has moved upstream: to the data you feed Google’s AI so it doesn’t waste your money, and to whether your brand even appears inside the AI answer, where your feed quality, your structured data, and your credibility decide whether you show up at all. That’s a different discipline — and most advertisers are still optimizing for a search results page that’s quietly disappearing. Let me show you what changed, what still matters, and what to actually do about it. The 2026 contradiction: better clicks, worse conversions Start with the numbers, because they tell a stranger story than “CPCs are rising.” WordStream’s 2026 benchmark data puts the cross-industry average Search CPC at roughly $2.96 in Q1 2026, up about 12% year over year — and other 2026 reports place the blended figure as high as $4.22 once high-cost verticals are weighted in. The spread is enormous: ecommerce sits near $1.16 a click while legal runs $6.75 to $8.58. None of that is new in shape; clicks have gotten more expensive every year since 2021. The interesting part is the contradiction underneath it. Across the industries WordStream tracks, click-through rates rose roughly 7.5% while conversion rates fell about 9% — declining in 13 of 14 verticals. Median CPA climbed around 12% to roughly $23.74. ROAS slipped about 10%. Read that chain slowly: the ads got better at earning clicks, and the pages got worse at converting them. That single pattern reframes the entire optimization conversation. When CTR is up and conversion rate is down, your problem isn’t your ad — it’s the match between what the click promised and what the page delivered. The bottleneck moved from the auction to the post-click experience. We’ll come back to that, because it’s where a lot of “expensive Google Ads” problems actually live. Why CPCs keep climbing — and why AI Overviews are part of it The auction mechanic hasn’t changed: Ad Rank is still bid × Quality Score × context. More advertisers, higher bids, simple. But there’s a newer pressure most analyses miss. AI Overviews — Google’s AI-generated answer panel — now satisfy a large share of informational and mid-funnel queries on the page itself. Independent research has been consistent and brutal here: a Pew Research Center study of tens of thousands of real queries found users clicked through only about 8% of the time when an AI Overview appeared, versus 15% without one. Seer Interactive measured organic click-through on AI-Overview queries falling from 1.76% to 0.61% between mid-2024 and late 2025. The knock-on effect for paid search is structural. When the AI answers the top-of-funnel question for free, the clicks that do survive are fewer and more decision-stage — which means more advertisers fighting over a smaller pool of high-intent traffic. CPCs go up not just from competition, but from compression. This is the same shift we documented for organic in GEO in the Philippines: why most Filipino e-commerce brands are already behind — the link economy is being replaced by an answer economy, and paid search is feeling the same gravity. So yes, optimize the auction. But understand that the auction is now playing on a smaller field. Quality Score still matters — but think of it as feeding the machine Quality Score remains the most reliable cost lever you control. Google’s own guidance and years of benchmark analysis line up: improving Quality Score from a 5 to an 8 cuts your effective CPC by roughly 30–37%. That’s not a rounding error. In a $6 legal click or a $40 supplement CPA, that’s the difference between profitable and pointless. What’s changed is the framing. Quality Score used to be a thing you gamed with tight single-keyword ad groups. That era is over. The modern structure is theme-based ad groups of 15–25 keywords, broad match paired with Smart Bidding, and ruthless negative-keyword hygiene. The old SKAG playbook now fights the algorithm instead of helping it. The three inputs haven’t moved — expected click-through rate, ad relevance, and landing page experience — but the way you earn them has. You earn expected CTR with responsive search ads that give Google real creative range. You earn ad relevance by grouping keywords by genuine intent, not by cramming unrelated services into one group. And you earn landing page experience with pages that mirror the ad’s promise and load fast, because Core Web Vitals still feed the quality signal. Think less “optimization” and more “feeding the machine the cleanest possible signal.” That mental shift is the whole game in 2026. The structural change you can’t ignore: AI Max is replacing Dynamic Search Ads If you take one operational action from this article, take this one. In April 2026, Google moved AI Max for Search out of beta — and confirmed it’s replacing Dynamic Search Ads. Starting in September 2026, campaigns still running DSA, automatically created assets, or the campaign-level broad match setting will be auto-upgraded to AI Max, and you’ll lose the ability to create new DSA campaigns across the UI, Editor, and API. This isn’t a feature you can sit out. AI Max is Google’s most