When to Build a Custom Website

Most businesses start on platforms — Shopify, Webflow, WordPress, Wix. They're fast, affordable, and handle 80% of standard needs.

But platforms have ceilings. At some point, the platform that helped you start becomes the thing holding you back.

Here are 7 signals that you've hit that ceiling.

Signal 1: Platform Fees Are a Material Business Cost

Platform transaction fees are the most common tipping point.

The math:

• Shopify Basic: 2% transaction fee (non-Shopify Payments)
• At $50,000/month revenue: $1,000/month in fees
• At $100,000/month revenue: $2,000/month in fees
• At $200,000/month revenue: $4,000/month in fees — $48,000/year

Custom development with Stripe direct costs 2.9% + $0.30 per transaction (Stripe's standard fee). No platform cut.

If platform fees are a line item you think about, it's time to model custom.

The number to check:

Monthly revenue × transaction fee rate × 36 months vs custom development cost

If the 3-year fee total exceeds your custom development cost, the build pays for itself in fees alone.

Signal 2: You're Building Workarounds, Not Features

This is the most common signal agencies see:

• Zapier automations 5 steps deep to approximate a feature
• Multiple apps that almost work together
• Developers spending time building "custom" on top of the platform
• Changing business processes to fit the platform instead of the other way around

If you're bending your business to fit your platform, the platform is wrong — not your business.

The test: List every automation, integration, and custom code you've added to your platform. If that list is longer than 10 items and growing, you're building a custom website on top of a website builder — and paying a premium for the privilege.

Signal 3: Performance Is Hurting Your Conversion Rate

Website speed directly impacts revenue. This is not a theoretical concern — it's measured.

Industry data:

• 1-second delay in load time → 7% reduction in conversions (Akamai)
• 2-second delay → 15% reduction
• 3-second delay → mobile users abandon at 53%

Platform performance reality:

• Shopify average First Contentful Paint: 2.5–4.5 seconds
• WordPress with typical plugins: 3–6 seconds
• Webflow: 2–4 seconds

Custom performance achievable:

• Next.js with edge deployment: 0.3–1.2 seconds

Calculate your performance revenue impact:

Monthly revenue × conversion rate improvement (10–25%) = additional monthly revenue

For a $100K/month store, improving conversion by 15% through performance = $15,000/month. A custom build pays for itself in 2–3 months of performance gains.

Signal 4: You Can't Build a Feature Your Business Needs

This is binary. Either your platform supports it or it doesn't.

Features platforms can't do:

• B2B pricing: different prices per account, negotiated pricing, tier-based discounts
• Multi-vendor marketplace: multiple sellers, vendor dashboards, commission splits
• Custom checkout flows: multi-step checkout with conditional logic
• Subscription + one-time combos: subscribe to a service + buy products in one cart
• Real-time inventory sync: multiple warehouses, real-time stock across channels
• Complex loyalty programs: points + tiers + referrals + custom rewards
• Custom authentication: SSO, B2B account management, team permissions

If a feature critical to your business model isn't available on your platform, that's the signal.

Signal 5: You're Paying Enterprise Prices for a Non-Enterprise Product

Shopify Plus: $2,300/month. BigCommerce Enterprise: custom pricing starting $1,500/month. These are significant costs — for a product that still has limits.

If you're spending $15,000–$25,000/year on a platform that still constrains you, that budget covers substantial custom development.

The calculation:

Annual platform cost ÷ 12 = monthly burn on someone else's infrastructure

If that number is over $500/month, custom development is worth modeling.

Signal 6: You Need to Own the Asset

This is the strategic signal most businesses overlook until it's too late.

Scenarios where ownership matters:

• Fundraising: Investors and VCs do technical due diligence. "We're on Shopify" is fine for early rounds, but as you scale, IP ownership matters.
• M&A: Acquiring companies value owned assets over platform rentals. A custom-built platform can be a genuine competitive moat.
• Enterprise partnerships: Large enterprise customers often conduct security audits. Platform-hosted sites have inherent limitations in compliance controls.
• Price risk: Platforms can and do raise prices. Shopify has increased prices multiple times. When you're doing significant revenue, a price increase is outside your control.

If any of these apply to your business in the next 2 years, the asset ownership question matters now.

Signal 7: Your Technical Debt Is Growing Faster Than Your Revenue

This is the advanced signal. It looks like:

• Your development team spends more time maintaining integrations than building features
• Your deployment process has grown complex because of platform workarounds
• New engineers take 2+ weeks to understand the system
• Every update to the platform breaks something in your customizations

This is architectural debt. It compounds. And it happens when you've built too much custom logic on top of a platform that wasn't designed for it.

The solution is a migration to a clean custom architecture — and the earlier you do it, the cheaper it is.

What to Do When You See These Signals

Step 1: Audit your current costs Add up: platform fees + transaction fees + app costs + developer workaround time. That's your real annual cost.

Step 2: Model the custom development cost Get quotes from 2–3 agencies. Compare against your 3-year current cost.

Step 3: Document your requirements List every feature you have, every workaround you've built, and every feature you can't build. That's your requirements document.

Step 4: Assess your timing Migration takes 3–6 months. Is there a low-traffic period? A natural product cycle milestone? The right timing reduces risk.

Step 5: Plan for migration SEO Every migration risks SEO ranking drops. Plan URL redirects, schema preservation, and Google Search Console monitoring from day one.

The Timing Rule

If you've seen 2 or more of these signals, it's time to start planning — even if you don't move immediately.

Custom development has a lead time. Planning now means moving in 3 months. Waiting until the pain is acute means an emergency migration at a premium price.

The businesses that regret going custom usually tried to cut corners. The businesses that regret staying on platforms usually wish they'd made the move 18 months sooner.