Do you actually need a new website? How to tell where the real problem sits

Two complaints usually turn up in the same meeting. The website gets traffic but very little of it turns into an enquiry. And nobody internally can change anything without raising a ticket and waiting a fortnight. Put those together and the conclusion writes itself: we need a new website.

Sometimes that is right. Often it is an expensive way to avoid a diagnosis.

A rebuild takes months, absorbs most of the marketing team's attention, and pauses the improvements you could have made to the site you already have while it is in progress. It also only fixes a specific set of problems. If your problem is not on that list, you will relaunch on time, on budget and on brand, and find the same conversion rate waiting for you on a much better looking page.

So before the budget conversation, it is worth being precise about what is actually failing.

A rebuild solves a narrow set of problems. Check yours is on the list

A new website is a genuinely good answer when the problem is structural. That usually means one of the following.

The story is wrong. Visitors cannot tell what you do, who it is for, or why it matters commercially, and no amount of button testing will rescue a proposition that is not there.

The structure is wrong. The way the site is organised reflects how your business is organised rather than how buyers look for things, so people cannot find the page that would have convinced them.

The platform genuinely blocks the work. The site cannot integrate with the systems you sell through, cannot be measured properly, or cannot be changed without a developer, and that limitation is now costing you commercial opportunities rather than just irritating people.

The site cannot carry what you need it to carry next. New markets, new languages, new products, a self-serve journey, a customer portal, a partner area.

Notice what is not on that list. A poor conversion rate is a symptom, not a cause. It appears in all four of the situations above, and in several others that a rebuild will not touch.

Where the conversion problem usually sits

Before you accept the diagnosis, work through these four possibilities in order. They are ranked by how often they turn out to be the real constraint, and by how cheap they are to test.

The traffic is arriving with the wrong expectation

Site-wide conversion rate is close to a meaningless number. It blends people who searched for your brand name with people who clicked a broad paid search advert while researching something adjacent, and the two behave nothing alike.

Split the number by source and intent, and the picture usually changes. Sometimes the site converts perfectly well for people who arrive knowing who you are, and badly for paid traffic that was promised something the landing page does not deliver in its first screen. That is a message match problem, and it is solved by changing the advert, the landing page or the targeting, not the whole website.

McAfee is a useful example of the difference. Consumers arriving from paid search and display were landing on pages that did not follow through on the promise in the advert. The work was research-led experimentation on those existing journeys, focused on aligning the advert with the page and simplifying the checkout, and it produced an 11% increase in conversion rate and a 16% increase in return on paid media spend. The gap between the advert and the page was the constraint, and closing it did not require rebuilding anything.

The journey asks for too much, too early

Most B2B sites are built for the small minority of visitors who are ready to talk to sales, and offer nothing at all to the majority who are still working out whether they have a problem worth solving.

If your only conversion point is "book a demo", your conversion rate is partly a measure of how many people arrived ready to buy. That is a demand and offer problem rather than a design problem. Look at whether form starts are high and completions are low (a friction problem), or whether nobody starts at all (a relevance or offer problem). They lead to completely different fixes.

There is external evidence for looking at the form itself rather than assuming the whole site is at fault. HubSpot's analysis of more than 40,000 landing pages found conversion rates tended to decline as fields were added, with more demanding inputs such as text areas and multiple dropdowns associated with particularly pronounced falls in conversion. The useful question is therefore not simply whether visitors converted, but where the commitment being asked of them became greater than their motivation to continue.

The enquiry is fine and the follow-up is not

This is the most commonly missed cause, because it sits outside the website and outside marketing's dashboard.

If enquiries are handed to sales through a form notification, an inbox or a manual export, some of them will not be worked, and the ones that are worked may be contacted days later. The site gets blamed for a revenue shortfall it did not cause.

The effect can be substantial. The original Lead Response Management Study analysed more than 15,000 web leads and 100,000 call attempts and found the odds of qualifying a lead were 21 times higher when the first attempt was made within five minutes rather than 30. Later Harvard Business Review research found companies responding within an hour were nearly seven times more likely to qualify a lead than those waiting another hour.

More recent data points in the same direction. InsideSales analysed 55 million sales activities involving 5.7 million inbound leads across more than 400 companies and found conversion rates were more than eight times higher when the first attempt happened within five minutes compared with leads first attempted between five minutes and 24 hours.

The precise multiplier will vary enormously by business, buying cycle and enquiry type. The operational point is harder to argue with: the value of the traffic you have already paid to generate can deteriorate while it is sitting in somebody's inbox.

So before you commission a rebuild, find out what proportion of enquiries from the last quarter received a human response within five minutes, one hour and one working day — and what happened to the rest.

You cannot see any of this clearly enough to judge

Plenty of rebuild decisions are made on numbers nobody actually trusts. Analytics says one thing, the CRM says another, paid platforms claim conversions that never appear as pipeline, and form submissions are counted differently depending on who is asking.

If your analytics and your CRM disagree about how many enquiries you received last month, that disagreement is the first thing to fix. It is far cheaper than a rebuild, and until it is fixed you have no reliable way of knowing whether a rebuild worked.

"Clunky to update" is rarely just a platform problem

The internal experience matters. A site that only two people can safely edit will always be out of date, and that has a real commercial cost: campaigns launch late, pricing pages stay wrong, and the team stops trying.

But slowness usually has more than one cause, and swapping the platform only addresses one of them.

Try this. Write down the last five things you wanted to change on the website. Note how long each one took from request to live, and where the time actually went.

If the delay was waiting for a developer to build a layout that does not exist yet, the problem is the absence of a component library, not necessarily the platform.

If the delay was approvals, the problem is governance, and it will follow you onto any new system.

If the delay was that nobody in the team is confident enough to publish, the problem is training and permissions.

If the delay was that the platform physically cannot do it, or the change is only possible by editing code, then the platform is genuinely the constraint.

Community is a reasonable illustration of the difference. The issue was not simply that the site looked dated. Their existing site did not connect properly to the marketing platform they were running campaigns in, load times were creating friction, and the journeys did not reflect the different buyers they were selling to. The rebuild included a reusable component library and around thirty pages of rewritten copy, so the marketing team could publish and track campaigns without going back to development every time. The internal use case was part of the brief, not an afterthought.

That is the standard worth applying. If you are rebuilding, be explicit about what the marketing team must be able to do unaided the week after launch, and treat it as a requirement rather than a hope.

Five things to establish before you sign off a rebuild

None of these takes longer than a fortnight to pull together, and together they usually change the shape of the project.

Conversion rate by source and by page type, not site-wide. Where exactly does performance fall away?

Form starts against form completions, plus the point in the journey where people leave. This distinguishes friction from irrelevance.

The gap between your analytics and your CRM. If the two numbers differ, quantify the difference and find out why before making a decision based on either.

Time from enquiry to first human contact, including the proportion reached within five minutes, one hour and one working day, and the proportion that never receives a response.

Your change log. What you tried to change in the last quarter, how long it took, and what actually blocked it.

If those five point at the same thing, you have a diagnosis. If they point in four directions, you have a sequencing problem, and a rebuild is likely to be the most expensive way to address it.

When replatforming genuinely is the answer

Sometimes the platform really is the ceiling, and continuing to optimise around it is the expensive option.

MirrorWeb's leadership recognised their existing website could no longer connect marketing activity to measurable commercial outcomes, and moved the site from Webflow onto HubSpot with a fully custom theme. The migration was scoped at thirteen weeks and delivered in eleven. Demo requests increased by 700% after the migration, and marketing performance became directly attributable for the first time.

Their VP of Marketing, Marissa Jambrone, described the decision this way: "I was apprehensive about moving our website to HubSpot, but Future Group totally changed my perspective. They made what felt like a risky move one of the best investments we've made."

What made that the right call was not the state of the old design. It was that a specific, identified limitation — the inability to link marketing activity to commercial outcomes — could not be removed any other way. That is the test. Name the limitation, and be able to explain why nothing short of a rebuild removes it.

A sequencing decision, not a yes or no

The choice is rarely rebuild or do nothing. In practice there are four moves, and most businesses need more than one of them in a particular order.

Fix the measurement first, because every other decision depends on it and it takes weeks rather than months. Then fix the handover into sales, because it is usually the cheapest place to recover revenue you have already paid to generate. Then test the changes you believe a rebuild would deliver, on the site you already have, because a handful of experiments will tell you whether the proposition or the layout was the problem. Rebuild when you have evidence about what to change, not just a conviction that something must.

There is a useful side effect to that order. Experiments run before a rebuild become the specification for it. Instead of relaunching and hoping, you launch a site built on things you already know work, and you have a trustworthy baseline to measure it against.

The question worth taking into the next budget meeting

"Is our website good enough?" has no useful answer. Every website could be better, and the case for rebuilding can always be made on aesthetics alone.

A better question is narrower: what commercial result needs to change, and what is the specific thing standing in the way of it changing?

If the honest answer is that the platform cannot support what the business is trying to do next, build. If the honest answer is that nobody is quite sure where the money is being lost, spend a fortnight finding out.

It is the cheapest fortnight you will spend all year, and it will either save you a rebuild or make the one you commission considerably better.

If the issue turns out to be less about the site itself and more about what happens to interest once you have it, it is worth exploring how FutureGroup approaches growing through digital.