What is the actual difference between the two?
Optimization changes the parts; a redesign changes the system. Optimization assumes the structure is sound and improves what sits inside it — headlines, calls to action, proof, page speed, form length. A redesign rebuilds the structure itself: navigation, page architecture, visual language, and often the platform underneath.
Both can fix a falling conversion rate, and both can waste money if applied to the wrong problem. The mistake is treating them as competing budgets rather than sequential steps. Most businesses should exhaust the cheap path before committing to the expensive one — and most redesigns should include optimization work as a matter of course, because a new site nobody has tuned is just a new baseline.
What can optimization realistically fix?
Optimization is the cheaper, faster and lower-risk path — and it fixes more than most businesses expect. If visitors arrive and leave without acting, the usual culprits are message clarity, buried calls to action, missing proof and slow pages. All four are fixable without touching your design system.
Start by watching five real sessions on a phone. You will see the friction faster than any dashboard will show it.
The ceiling matters, though. Optimization can lift a 1.2% conversion rate to 2.5% over a few months. It cannot make a site built for a desktop era feel credible on a phone, and it cannot change what the site sells. When you hit that ceiling, more optimization produces diminishing returns and rising frustration.
- Rewrite the hero so it says what you do, for whom, and why you're different
- Add a visible next step at every scroll depth
- Move testimonials and guarantees next to the point of decision
- Compress images and remove render-blocking scripts
How do you run a cheap optimization sprint?
Block two weeks and work in order. Day one: watch five real sessions on a phone, using recordings rather than a dashboard. Day two: write down the single most common reason people leave. Days three to five: rewrite the hero, the primary call to action and the proof block on your top three pages.
Week two is technical. Compress images, defer non-critical scripts, preload the hero asset, and check that your enquiry form works on a cheap Android phone over 4G. Then measure the same conversion rate before and after over four weeks — not three days.
A sprint like this costs days rather than months, and it usually produces the single largest improvement available to a site that has never been deliberately optimized. Run it before deciding a redesign is necessary, because the result changes the decision.
What are the signs your platform is the real problem?
Platforms rarely fail loudly. They fail by making everything slightly harder: every new page needs a developer, every integration is a plugin you hope still works, and the page builder outputs markup that no amount of tuning will make fast.
The signal is the ratio of ideas to possible changes. If fewer than three of your top ten improvement ideas are achievable without a rebuild, you are maintaining a constraint rather than a website.
- Publishing a page requires a developer
- Core Web Vitals fail even after optimization work
- Integrations depend on plugins nobody maintains
- The mobile experience cannot be fixed without rebuilding templates
- Security updates arrive faster than you can apply them
What does each path actually cost?
Optimization typically runs a fraction of a rebuild — days or weeks of focused work rather than months. A redesign of a 10-page marketing site usually lands between $6,000 and $20,000 in 2026 once content, migration and SEO preservation are included.
The hidden cost in redesigns is not the build. It is the migration: redirects, rankings, analytics continuity and the institutional knowledge embedded in pages nobody remembers creating. Budget that separately from the build — it is where redesign quotes most often underdeliver, and a quote that lists it explicitly is one you can trust.
How do you decide when both paths are close?
When the case is genuinely marginal, run the cheap path first and treat it as evidence. Two weeks of optimization gives you a real conversion number on the existing site. If that number is close to your target, stop. If it barely moves, you have proved the problem is structural — and you enter the redesign with a baseline to beat.
There is also a middle path worth naming: a partial rebuild. Keep the URLs, the content and the technical foundations that work, and rebuild the templates and design system around them. It costs less than a full redesign and preserves far more SEO equity than a clean-slate approach.
How do you redesign without losing your rankings?
Plan the migration before you design a single page. Export your current URLs and their traffic, map every one to a destination, and write redirects for anything that changes. Keep your highest-performing pages' copy and structure where it works — a redesign is not a licence to rewrite everything.
Then launch quietly, watch Search Console for two weeks, and fix crawl errors within days rather than months.
Those first two weeks are the highest-risk period in the project. Compare impressions and clicks page by page against the previous period. Most post-redesign dips recover within a month — but only if someone is watching.
What should you do in the first week after launch?
Verify the plumbing before you celebrate. Submit the new sitemap, check that robots.txt is not blocking anything, confirm canonical tags point to the right URLs, and test every form and payment flow yourself on a phone.
Then resist the urge to keep changing things. A redesign needs a stable four weeks to produce a readable conversion number. Tweaking the hero every three days during that window makes the result uninterpretable, and you end up arguing about opinions instead of data.