A new VP of marketing inherits a website, spends two weeks clicking around it, and arrives at the same conclusion almost everyone does: this needs to be rebuilt. The pages feel slow. The admin is a mess. Half the templates look like three different companies made them. The instinct is to burn it down and start clean.
That instinct is usually wrong, and it is expensive to be wrong about. The site feeling broken is real. The diagnosis (the platform failed, so we need a new one) is the part that rarely survives an audit. Most marketing sites at 20-200 person B2B SaaS companies do not have a platform problem. They have an ownership problem wearing a platform costume. The Website Ownership Gap is the long version of why. This is the part where that gap gets misdiagnosed as a rebuild.
The Rebuild Reflex
The rebuild proposal almost never gets framed honestly. Nobody writes a deck that says “we neglected this site for three years and now it shows.” The deck says “the platform is holding us back,” because that framing assigns the failure to a vendor instead of to the organization. It is a more comfortable story, and comfort is doing more work in the decision than diagnosis is.
I have advised against more rebuilds than I have led. The pattern that triggers the conversation is consistent: the site is slow, the tracking is untrustworthy, and the marketing team is exhausted from waiting on engineering for every change. Each of those is a real problem. None of them is a platform problem, and a rebuild fixes none of them durably. You can rebuild a site on the best stack in the world and re-create all three within a year if the thing that produced them (nobody accountable for the site between launches) is still true on launch day plus one.
Think of it like a musician blaming the guitar. A great instrument in untrained hands still sounds bad, and a new one does not teach you to play. The rebuild buys a new guitar and changes nothing about who is holding it.
What a Rebuild Actually Costs
Start with the number everyone quotes and most people understate. In 2026, an agency breakdown of B2B SaaS redesigns priced them from $15,000 to well over $150,000 depending on scope, with typical timelines of 12 to 16 weeks and enterprise builds with content migration running four to six months[1]. A separate 2026 agency analysis of redesign costs lands in the same five-to-six-figure territory[2]. Two independent agency estimates, same order of magnitude. The rebuild is a six-figure-capable, multi-quarter commitment.
The dollar figure is the cost people budget for. It is not the cost that hurts.
The cost that hurts is the success rate. In 2023, the research firm SoftwareReviews reported that 80% of website redesigns fail to achieve their full potential, attributing the gap to a disconnect between what the business wanted and what its buyers actually needed[3]. Read that against the price tag. The most likely outcome of a six-figure rebuild is a site that looks newer and performs about the same, because the rebuild reorganized the pixels without resolving the buyer-needs question that a rebuild was never going to answer on its own.
The Risk Nobody Prices In
Here is the cost that almost never makes it into the rebuild deck: a rebuild is a migration, whether or not anyone calls it one. New templates, new URLs, new information architecture, new content model. To search engines, that is a site migration, and site migrations are dangerous.
In 2025, a study of 892 domain migrations found the median site took roughly 523 days to recover its prior organic traffic, and about 17% of domains never fully recovered even after 1,000 days[4]. That is not a worst case. That is the median. The typical migrated site spends close to a year and a half clawing back rankings it already had, and nearly one in five simply does not get all of them back.
The disaster cases are louder. When WooCommerce migrated its commercial site from woocommerce.com to woo.com in late 2023, public visibility data showed the domain shedding the large majority of its organic visibility almost immediately[6]. If a company that builds e-commerce software for a living can lose that much in a domain move, the assumption that your rebuild will be the clean exception deserves more scrutiny than it usually gets.
This is the math that makes casual rebuilds reckless. You are spending six figures and six months for a coin-flip improvement, while putting years of accumulated search equity on the table. The organic traffic you have today is an asset you already paid for. A rebuild risks it for a redesign that, four times out of five, will not reach its potential anyway.
Why Does the Site Feel Broken?
If the platform is rarely the problem, what is? In almost every case I have audited, the answer is the workflow around the site, and the absence of anyone whose job is to hold that workflow to a standard.
A marketing team I worked with ran a few events a month: webinars, partner sessions, regional roadshows. The process for shipping each one was to find the most recent event page, duplicate it, change the title and date, and publish. Reasonable, in isolation. Across a few hundred pages and a few years, it became a slow-motion disaster. Old logos from two rebrands ago survived on pages nobody revisited. CTAs used phrasing the company had retired. A third of the footers still linked to a sunset product. Color values drifted because someone had hand-coded a hex in 2022 that never got updated downstream.
The site looked nothing like the brand. The obvious read was “this needs a rebuild.” The actual problem was the cloning workflow, and the fix was structural, not a teardown: one event content type, one template, brand tokens pulled from a single source. Add an event, fill in five fields, publish. We backfilled the existing pages into that structure and the inconsistencies disappeared in a single deploy. The platform never needed to change. The way work flowed through it did.
The same dynamic shows up as platform bloat. Another site I cleaned up had accumulated 12 custom post types and 40 plugins, half of them doing nothing. Some were installed for an experiment three vendors ago. One was still firing database queries on every page load for a feature that had not existed since 2022. The admin felt slow, and “slow admin” reads as “old platform, time to rebuild.” Removing what nobody used made everything faster, no rebuild required. Nobody had ever been assigned to subtract, because new work always had a budget and old work always had inertia.
This is the heart of it. The website drifts the way anything drifts when the answer to “who owns this” is a shrug. A rebuild resets the drift to zero for exactly as long as it takes the same ownerless workflow to start it again.
When Is a Rebuild Actually the Right Call?
The case against casual rebuilds is not a case against all rebuilds. Sometimes the platform genuinely is the constraint, and pretending otherwise wastes more money than a rebuild would.
The real triggers are specific and they announce themselves through a named limitation, not a general mood. The CMS structurally cannot represent the brand or the content types the business now needs. A headless migration is overdue because performance is capped by the current architecture and you have already exhausted the optimization headroom. The data model cannot hold what the marketing motion requires and every workaround makes it worse. An acquisition or rebrand forces a consolidation that genuine surgery cannot reach.
The test is one question: can anyone on the team name the specific structural thing a rebuild would unlock that focused ownership work could not? If the answer is concrete (we cannot build the use-case pages the new ICP needs because the page model does not support them), that is a real rebuild trigger. If the answer is some version of “it just feels like time,” that is the ownership gap talking, and a rebuild is the most expensive way to quiet it temporarily.
The Alternative Almost Nobody Budgets For
Between “live with the broken site” and “spend six figures starting over” is the option most teams skip, because it is unglamorous and has no launch party: own the site and fix it in place.
Most B2B SaaS marketing sites on WordPress, Webflow, or HubSpot CMS can be brought to current standards in two to four months of focused ownership work. A performance pass. A tracking rebuild. Template consolidation. Content surgery on the pages that drifted. The result is usually 80% of what the rebuild would have delivered, with none of the migration risk, no SEO equity loss, and no quarter of operational disruption. The decision between the two is worth making deliberately rather than by reflex, which is why it deserves its own dedicated framework rather than a gut call in a planning meeting.
The contrast is stark once you lay the two paths side by side.
| Dimension | Full rebuild | Own and fix in place |
|---|---|---|
| Timeline | 12 to 16 weeks, up to 6 months for enterprise | 2 to 4 months, shipping fixes weekly |
| Cost | Five to six figures | A fraction of a rebuild; mostly time |
| SEO risk | High: median 523 days to recover after a migration | Minimal: no URL or template churn |
| Reversibility | Low: a big-bang launch is hard to roll back | High: every change is incremental and testable |
| Right when | The audit names a structural constraint | The problems are diagnosed and fixable |
The redesign treadmill makes this harder than it should be. The idea that companies should rebuild every two or three years is remarkably persistent and remarkably unsourced. Even the agencies that cite the three-year rule cannot point to where it comes from[5]. It is folklore that happens to be good for the people who sell rebuilds. The site does not have an expiration date. It has a maintenance requirement, and the teams that meet it rarely need the teardown.
The Rebuild Is Usually Cost Number Four
The cleanest way to see the rebuild reflex is to place it where it belongs in the sequence. An unowned marketing site generates four costs: conversions lost to slow pages, ad spend wasted against broken tracking, marketing velocity eaten by the engineering backlog, and finally the rebuild that gets proposed to escape the first three. That last one is the fourth cost of an unowned site, and it is the one teams notice because it is the only one with a budget request attached.
The first three are the actual problem. The site is slow because nobody runs the business case for site speed as a monthly metric. The numbers are untrustworthy because the Analytics Trust Gap went unaudited for a year. The team is slow because every change waits in the product backlog. A rebuild does not resolve any of those. It resets them and waits for them to return, which they will, on the same schedule, for the same reason. The pattern is named in full in the hidden cost of website neglect.
A rebuild is the right answer to a structural constraint you can name. It is the wrong answer to a site that feels broken because no one has been responsible for it. The difference is not a matter of taste or budget. It is a matter of whether anyone has actually looked. Most teams have not, which is exactly why the rebuild keeps getting proposed and the gap keeps getting paid for.
Sources
- Line and Dot Studio, What a B2B SaaS Website Redesign Actually Costs in 2026 – June 2026 agency pricing breakdown; B2B SaaS redesigns $15,000-$150,000+ by scope; typical timeline 12-16 weeks, enterprise content-migration builds 4-6 months ↩
- Webstacks, How Much Does a Website Redesign Cost? (2026) – April 2026 agency analysis of redesign costs; mid-range redesigns ~$10,000-$40,000, enterprise-grade $40,000-$80,000+; cited as a corroborating same-territory cost reference ↩
- SoftwareReviews (Info-Tech Research Group), 80% of Website Redesigns Fall Short of Full Potential – July 2023 research release; 80% of website redesigns fail to achieve their full potential, attributed to a business-versus-buyer-needs disconnect; sample size not disclosed ↩
- Search Engine Journal (Dan Taylor), How Long Should an SEO Migration Take? – January 2025 study of 892 domain migrations (data collected Oct 22, 2024); median 523 days to regain prior organic traffic; ~17% never fully recovered after 1,000 days ↩
- Walker Sands, Why Companies Redesign Their Websites Every Three Years (Or Do They?) – Agency analysis noting the commonly cited three-year redesign cadence has no identifiable primary source ↩
- Totally Digital, SEO Migration Failures: Examples of Traffic Declines – Case study citing public visibility data; WooCommerce’s late-2023 woocommerce.com to woo.com move shed the large majority (~90%) of organic visibility almost immediately ↩
Seeing these patterns at your company?
Book a free WebOps Diagnostic. I'll review your site before the call and share specific observations.
Book a Free Call →Frequently Asked Questions
Independent 2026 agency estimates price a B2B SaaS redesign from roughly $15,000 to well past $150,000 depending on scope, with typical timelines of 12 to 16 weeks and enterprise builds with content migration running four to six months. The dollar figure is only part of it. A rebuild also costs a quarter of marketing velocity, the risk of losing search rankings during migration, and the opportunity cost of the iterative work you did not do instead. For most 20-200 person companies, the all-in cost is higher than the budget line suggests.
It can, badly, and the risk is routinely underpriced. A 2025 study of 892 domain migrations found the median site took roughly 523 days to recover its prior organic traffic, and about 17% never fully recovered even after 1,000 days. A rebuild that changes URLs, templates, and information architecture is a migration whether you call it one or not. Without disciplined redirect mapping and parity testing, you can lose rankings that took years to earn and may not get back.
Run a focused audit first. If you cannot name the specific structural thing a rebuild would unlock that ownership work could not, you probably do not need one. Genuine rebuild triggers are concrete: the CMS cannot represent the brand, a headless migration is overdue for performance reasons, or the data model fundamentally cannot hold your content types. Generalized frustration (the site feels old, the admin is slow, dev never has time) is a symptom of the ownership gap, not a case for starting over.
Because the first three costs of an unowned site (slow pages, broken tracking, team friction) stack until starting over feels like the only clean option. A rebuild is emotionally satisfying. It is visible, it has a budget, and it lets nobody name that the failure was operational rather than technical. The redesign-every-few-years instinct is folklore, not data. Even the agencies that cite a three-year rule cannot point to its source.
