When a website starts feeling outdated, many businesses instinctively propose a complete redesign: scrap what exists and launch a brand new site from top to bottom on a specific date. It is a legitimate option in some cases, but not the only one, and certainly not always the safest one. The alternative (incremental redesign, changing parts of the website progressively while the rest keeps working) is usually the lower-risk option for businesses that already get real traffic and sales through their current website.
What a "big bang" redesign means
A big bang redesign involves building the new website in parallel, usually in a separate environment, and launching it all at once on a specific date, replacing everything that came before in one go. It has an obvious advantage: the final result is coherent from day one, with no intermediate phases that might look half-finished. The risk is just as obvious: if something goes wrong (an important page loses ranking, a purchase flow converts worse than the old one, a technical bug not caught in time), the impact hits the whole business at once, with no room for gradual reaction.
What an incremental redesign means
Incremental redesign means gradually renewing parts of the website (a section, a page template, a specific flow), measuring the impact of each change before moving to the next. It is slower to reach the full final result, but it lets you catch and fix problems at a small scale, before they affect the entire site, and it always keeps a working version generating business while the next improvement is underway.
The real risk of the big bang: losing SEO rankings overnight
One of the most underestimated dangers of a full redesign is the impact on search engine rankings. Changing URL structure, removing content that had ranked for years, or altering the site's architecture without carefully planning redirects can cause Google to "lose" pages that had been bringing in traffic for a long time, and that traffic does not come back overnight, even if the new design is objectively better. An incremental redesign, by changing fewer things at once, drastically reduces this risk.
When the big bang approach genuinely makes sense
It is not all downsides for the all-at-once approach. It clearly makes sense when the current website is built on an outdated technical foundation that prevents any reasonable incremental improvement (a discontinued platform, unmaintainable code), when there is a complete rebrand requiring total coherence from launch, or when current traffic is so low that the risk of losing rankings is practically irrelevant. In those scenarios, trying to go "little by little" on a broken foundation can end up costing more in the long run than starting from scratch.
When incremental is clearly the more sensible option
For a business with established traffic, real sales depending on the current website, and search rankings that took years to build, incremental redesign is almost always the more prudent choice. It lets you keep selling while improving, learn from each change before applying the next one, and avoid the most feared scenario: launching a new website and discovering two weeks later that something important converts worse than before, with no easy way back that does not also lose the new work done.
How to prioritise what to change first in an incremental approach
Sensible prioritisation starts with the pages that get the most traffic or have the most sales impact (where an improvement, however small, has the greatest proportional effect) and with the most obvious problems identified through analytics or real user testing (an abnormally high abandonment rate at a specific step, a page that loads much slower than the rest). Starting with what matters least "because it is easier" is a common mistake that delays the improvements that would actually move the needle.
The emotional cost of a gradual redesign: accepting temporary inconsistency
A side effect of the incremental approach rarely anticipated is that, for months, the website will have parts of the old design and parts of the new one coexisting. For some businesses that temporary visual inconsistency creates discomfort ("it looks half-finished"), even though avoiding the risk of an all-at-once change is precisely the goal. It is worth consciously anticipating and accepting this intermediate phase before starting, so the plan is not abandoned halfway through out of pure aesthetic impatience.
Communicating the redesign plan internally to avoid misaligned expectations
Regardless of the chosen approach, a common mistake is failing to clearly communicate to the rest of the team (sales, customer service) what is changing and when. If the sales team keeps showing old screenshots in their presentations while the website has already changed, or if customer service gets enquiries about a feature that no longer exists without knowing it was removed, internal misalignment creates avoidable friction that has nothing to do with the chosen technical approach and everything to do with the communication around it.
The role of A/B testing within an incremental redesign
One of the less-discussed benefits of incremental redesign is that it allows running real A/B tests before fully committing to a change: showing the new version of a section to only part of the traffic, measuring the real impact, and deciding with data whether it is worth rolling out to the whole site. This kind of validation is practically impossible in a big bang approach, where the entire site changes at once with no direct point of comparison.
When to stop an incremental redesign that is not delivering the expected results
An additional advantage of the incremental approach is the ability to pause or revert a specific change that is not working, with no need to undo the whole project. Deciding in advance which metrics will be watched after each change, and within what timeframe the call is made to continue or revert, prevents a problematic change from sticking around out of sheer inertia or the emotional cost of admitting it did not work as expected.
The budget factor: how spending is distributed in each approach
A big bang redesign concentrates spending into a defined period, which makes it easier to budget in one go but requires having that capital available all at once. An incremental redesign spreads spending over a longer time, which makes cash flow planning easier for businesses unable to absorb a large one-off outlay, even though the total accumulated cost is not necessarily lower.
Documenting the decisions made at each phase of an incremental redesign
Since an incremental redesign stretches out over time, it is easy to lose track of why a specific decision was made in an earlier phase, especially if someone on the team changes along the way. Keeping a brief record of what was changed, when and why at each phase avoids having to reconstruct that reasoning from memory months later, and makes it easier to assess with perspective whether the redesign as a whole is moving in the right direction.
The end customer's perspective: they rarely care which approach was chosen
It is easy for the internal discussion over which approach to choose to overshadow a more important question: what the end customer actually perceives. In most cases, the customer does not care whether the redesign was incremental or all at once, the only thing they notice is whether the website works better or worse than before. Keeping the focus on that perceived outcome, above the internal methodological debate, helps keep sight of the project's real goal.
Frequently asked questions
How long does a full incremental redesign process usually take?
It depends on the size of the website and the resources available, but a full incremental redesign commonly spans anywhere from several months to a year, compared to the shorter, more defined timelines of a big bang redesign planned around a fixed launch date.
Is incremental redesign always cheaper than the big bang approach?
Not necessarily in total cost, but it usually spreads the spending out better over time and reduces the risk of having to fix a full launch on the fly when it does not perform as expected, which can end up costing more in the long run.
Can I mix both approaches depending on the part of the website?
Yes, and it is actually common: renewing a specific section with serious problems all at once (the entire checkout process, for example) while the rest of the website gets updated incrementally is a perfectly reasonable strategy.
How do I protect my SEO rankings if I decide to do a full redesign?
The essential thing is to keep, whenever possible, the same URL structure, and when it does change, implement correct redirects from every old URL to its new equivalent, as well as preserving content that was already generating traffic instead of simply removing it.
How do I know if my website needs a full redesign instead of an incremental one?
If the current technical foundation prevents reasonable changes without disproportionate effort, or if the platform it is built on no longer receives support or updates, those are clear signs a full redesign may be unavoidable in the medium term.
Who should decide which approach to take, the business or the agency?
It should be a joint decision: the business knows the real risk it can take on (how dependent it is on current traffic, how much financial room there is for an adjustment period), and the agency or technical team knows the real limitations of the current foundation. Neither side should decide this alone.
How do I communicate an incremental redesign spanning several months internally?
With brief, regular updates to the rest of the team (what has changed, what is coming next), avoiding a situation where the information stays only with whoever runs the project directly, creating misalignment with sales or customer service.
Can I run A/B tests in a big bang redesign?
It is much harder, since the entire site changes at once with no direct control version to compare against. It is one of the strong arguments in favour of the incremental approach for businesses that value data-based validation before fully committing.
Which approach spreads spending out better over time, incremental or big bang?
Incremental, by spanning phases, makes it easier to spread spending over a longer period, which can be more manageable for businesses unable to absorb a large one-off outlay, even though total accumulated cost is not necessarily lower.
Is it worth documenting the decisions made at each phase of an incremental redesign?
Yes, especially in long projects where someone on the team may change along the way. A brief record of what was changed and why avoids having to reconstruct that reasoning from memory months later.
Does the end customer actually care whether the redesign was incremental or all at once?
In most cases, no: the only thing they perceive is whether the website works better or worse than before. Keeping the focus on that perceived outcome helps keep sight of the project's real goal above the internal methodological debate.