.png)
There is a particular kind of website that comes out of a long approval process involving many stakeholders. It is inoffensive. The colour palette is acceptable to everyone. The messaging is broad enough to include every service the company offers. The photography is professional without being distinctive. Nobody objects to any of it. Nobody is particularly moved by any of it either, which is a problem, because the people who are not moved by it are the visitors the site needs to convert into customers.
Design by committee does not produce bad websites by accident. It produces a specific kind of mediocrity, predictably and consistently, through a process that feels thorough and collaborative but is optimising for the wrong thing. Understanding the mechanism is the first step toward building something that works.
The committee is solving the wrong problem
A design approval process is intended to catch errors and ensure the work reflects the business accurately. What it actually does, in most cases, is apply internal political logic to decisions that should be governed by external commercial evidence.
When multiple stakeholders review a design, each person's approval is based on whether the work meets their own expectations and preferences. The marketing manager has views on brand alignment. The sales director wants every service category represented so their team has something to reference. The managing director is attached to a particular piece of copy because they wrote it three years ago and it has always been on the website. None of these perspectives is inherently wrong. None of them is asking the right question: what does a visitor who has never encountered this business need to see in order to move forward?
The design that passes committee review is the design that generates the fewest objections. This is not the same design that generates the most enquiries or the highest conversion rate. Those outcomes require choices that some stakeholders will disagree with: prioritising one message over others, removing services from the homepage that have lower commercial priority, writing copy that sounds specific rather than comprehensive. These decisions make somebody uncomfortable. In a committee approval process, discomfort is veto power.
How committees select for mediocrity
The mechanism is worth understanding in detail because it explains why committee-driven projects follow such a predictable pattern.
When a design direction creates a strong reaction, it creates a strong reaction across the room. Half the people in the approval meeting might love a bold approach. The other half find it too risky. The person who objects loudest tends to carry the review, not because their objection is well-founded but because design decisions rarely have clear evidence behind them at the point of approval. Without evidence, the safest move is to defer to the most risk-averse voice. The design gets pulled back toward something everyone can live with.
This process repeats across every element of the site. The hero image that was distinctive gets replaced with something more conventional. The headline that made a specific claim gets softened to accommodate a wider audience. The CTA that was direct gets qualified with additional language because one stakeholder felt it was too pushy. Each individual modification seems reasonable in isolation. The cumulative effect is a site that looks like every other site in the category, says the same things, and gives visitors no particular reason to prefer it.
The irony is that the process feels rigorous. Multiple rounds of review, detailed feedback, careful revision. The effort invested in the approval process creates a sense that the outcome has been thoroughly validated. It has been validated against internal preferences. It has not been validated against the behaviour of actual visitors, which is the only validation that matters commercially.
The specific damage committees do
There are patterns that appear in almost every committee-driven website build. They are worth naming specifically because they are often invisible to the people inside the process.
Homepage messaging that tries to say everything. When multiple stakeholders contribute to the hero section, each one adds their priority. The result is a homepage that lists six things in the hero instead of leading with one clear, compelling reason to stay. Visitors make the decision to leave a page in seconds. A hero that makes them work to understand what the business does is a hero that does not convert.
Navigation structured around the organisation chart. Committees tend to organise site navigation the way the business organises itself internally: by department, by service line, by product category. This makes sense to the people inside the business. It rarely reflects how a customer thinks about their own problem. A visitor who needs to improve their ecommerce conversion rate is not looking for a tab called "Digital Strategy." They are looking for something that matches the problem they are trying to solve. Navigation built around the company's internal structure forces visitors to translate, and most of them do not bother.
Safe photography and design that communicates nothing distinctive. When a design goes through multiple rounds of approval, visual risk gets squeezed out. The striking image gets replaced by the broadly acceptable one. The distinctive colour choice gets moderated toward something more familiar. The design becomes technically competent without communicating anything that differentiates the business from its competitors. For NZ businesses competing in categories where several players look roughly similar online, undifferentiated design is an invisible cost.
Copy written to avoid dissent. When sales, marketing and leadership all have approval rights over website copy, the resulting text reflects all of their priorities simultaneously. The copy that results is often accurate, comprehensive and completely ineffective at moving a visitor toward action. It describes the business from the inside out rather than addressing the visitor's concern from the outside in. It uses the language the business uses to talk about itself rather than the language a customer uses to describe their problem.
Why the people in the room are the wrong validators
The stakeholders in a website approval process are the worst possible audience to evaluate whether the design will work. Not because they lack intelligence or genuine investment in the outcome. Because they know too much.
A senior person who has been in the business for ten years cannot replicate the experience of a visitor encountering the company for the first time. They already understand what every service is and why it matters. They already know what differentiates the business from competitors. They read the website through a lens of existing knowledge that no visitor will ever have. When they say a page feels clear, they mean it is clear to them. That is not a useful signal.
The feedback that actually improves website performance comes from people who have no prior relationship with the business: recorded sessions of first-time visitors, user testing with the actual target audience, analysis of where visitors drop off and what they were looking for when they left. This kind of evidence does not require consensus or committee sign-off. It requires looking at what people actually do, not what the people inside the business think people will do.
Structuring projects so stakeholder input is useful
The goal is not to exclude stakeholders from the process. Their input on brand accuracy, factual content and business priorities is genuinely valuable. The goal is to structure the process so that input is applied to the decisions stakeholders are qualified to make, and not to the decisions where visitor behaviour is the relevant evidence.
Stakeholders should own: factual accuracy, brand representation, the completeness of the service offering, legal and compliance requirements. These are decisions where internal knowledge is the correct basis for judgment.
Design and conversion decisions should be governed by evidence: what the data shows about how visitors currently behave on the site, what the audience research reveals about how customers make decisions in this category, and what the performance benchmarks are for comparable sites. Where there is no existing data, the right approach is to test rather than debate.
The practical implication is that approval processes need to be scoped. Asking stakeholders to approve content accuracy is appropriate. Asking them to approve conversion copy, hero message hierarchy and CTA placement is asking them to govern decisions where they do not have relevant evidence. This is how projects end up with beautiful design rationale documents and homepage messaging that fails to generate enquiries.
What good looks like
The NZ businesses that get strong commercial results from their website projects have almost always restricted stakeholder approval authority to the domains where internal judgment is valid, and put performance decisions in front of the people with evidence to make them.
The Webflow site Skyrocket built for a NZ premium sauna brand converts enquiries at 4.6% against a category benchmark of around 2.7%. The messaging hierarchy on that site was not arrived at by committee. It was arrived at by understanding what a visitor considering a premium sauna purchase actually needed to understand before making contact, and structuring the page around that decision process. Some of those structural choices would have been softened or reversed in a standard approval process. They were not softened, and the conversion rate reflects that.
A website project is not a brand exercise where the goal is internal alignment. It is a commercial asset where the goal is performance against a measurable outcome. The businesses that treat it as the former tend to produce sites that their team is proud of and their analytics do not support. The businesses that treat it as the latter tend to produce sites that occasionally make someone in the approval meeting uncomfortable and consistently move the commercial needle.
The measure of a good website is not whether everyone in the room agreed to it. It is whether the people it was built for take the action it was designed to prompt.
