Traffic is up. Enquiries are flat. Someone in the meeting suggests increasing spend, because traffic is the one thing you can reliably buy.
It is worth pausing before that budget gets approved, because more visitors will not change the outcome if the reason people leave has nothing to do with how many of them there are. A site that fails to convert is rarely one problem with one owner. It is usually three problems that look identical in a dashboard and need completely different money spent on them.
Your conversion rate is hiding three different problems
If visitors arrive and leave without acting, the cause almost always sits in one of three places.
The people arriving are not in the market for what you sell. That is a positioning, messaging and targeting problem, and no amount of design work will fix it.
The people arriving are the right people, but they cannot make progress towards a decision. That is a site, content and journey problem.
Or the journey is working better than your reporting shows, because you are measuring the wrong event, or averaging together visitors who were never going to buy with visitors who were. That is a measurement problem, and it disguises the other two.
More traffic makes the first problem more expensive, the second problem more visible and the third problem harder to see. Which is why the sequence matters: understand which one you have before you spend anything on fixing it.
Conversion rate averages two things that should never be averaged together
A single conversion rate combines the intent of the people arriving with the effectiveness of what happens when they do. Those two things move independently, and often in opposite directions.
When you increase reach, you almost always widen intent. New markets, broader keywords, cooler audiences and awareness activity all bring in people who are earlier in their thinking or not buyers at all. The site did not get worse. The mix changed. A falling conversion rate in that situation can sit alongside a genuinely improving commercial picture.
The reverse is also true, and less often noticed. A stable conversion rate can hide a journey that has quietly deteriorated for your most valuable visitors, because a growing volume of easy, high-intent branded traffic props up the average.
The scale of that effect is easy to underestimate. First Page Sage's 2026 business-to-business benchmarks put average conversion at 2.6% for organic search against 1.5% for paid search, 0.9% for paid social and 0.3% for display. The same dataset reports organic conversion of 3.6% for companies selling to smaller businesses against 1.6% for those selling to enterprise. A business deliberately moving upmarket could therefore watch its headline conversion rate fall while its commercial strategy works exactly as intended.
Which is why there is no useful answer to "what should our website conversion rate be?" without first asking: for which audience, arriving from where, and converting into what?
So before drawing any conclusion, check three numbers that do not average intent away:
- Conversions in absolute terms, not just the rate. If enquiries are rising while the rate falls, you are buying broader intent, which is a budget question rather than a website question.
- Enquiries per £1,000 of media spend, by channel. This tells you whether more traffic is actually affordable.
- Value per visitor, or revenue per visitor if you sell online. It is the honest measure of whether the journey is getting better at its job.
Problem one: the wrong people are arriving
This is the diagnosis most teams reach for last, partly because it implicates the marketing strategy rather than the website.
The signals are reasonably distinctive. Traffic grows but the shape of it does not match anything you sell. Content pages attract strong engagement that never leads anywhere near a product or service page. Sales report that the enquiries coming through were never viable. Non-branded search brings volume while branded search brings almost all of the conversions. Or a meaningful share of your visits turn out to be candidates, suppliers, existing customers and competitors, none of whom were ever going to fill in a form.
Two causes sit underneath that.
The first is targeting. Broad match keywords, cheap inventory and audiences chosen for reach rather than fit all buy attention from people who cannot buy. That is a media problem and it is usually fixable within a quarter. It is also the subject of a separate piece on what to check before cutting paid media spend.
The second is harder. What the market thinks you do may no longer match what you actually sell. Businesses tend to outgrow their own language, so the words that made sense when the company was founded keep attracting the enquiries the company has moved past. If your positioning describes a narrower or older version of the business, the site will keep drawing in the wrong buyers and the right ones will not recognise themselves.
thingtrax is a clear example. The business had grown into enterprise manufacturing, but its materials had accumulated piecemeal and said different things in different places, so buyers could not form a consistent picture of what the company was or who it served.
The fix was one brand system rather than more traffic: a single, coherent account of the business that made its enterprise credibility legible. No amount of additional spend would have solved that, because the people arriving were responding accurately to what they had been told.
A blunt diagnostic question: what proportion of the people arriving on your site are even inside your addressable market? If nobody can answer that with evidence, the conversion rate is not the number to be arguing about.
Problem two: the right people arrive and cannot make progress
Assume the traffic is fine. The next question is whether a qualified visitor can actually get anywhere.
Most sites answer two of the four questions a buyer needs answered, and skip the two that decide the outcome:
- What is this?
- Is it for someone like me, in my situation, at my size?
- Why you rather than the alternative I am already considering?
- What happens if I get in touch, and what am I committing to?
The first and last are usually covered. The middle two are where the decision is actually made, and they are the ones most often replaced by capability lists. A page that describes everything a company can do puts the work of relevance onto the reader. Senior buyers do not do that work. They leave.
There is a second, less obvious version of this problem: the friction that costs the most is rarely the thing everyone argues about in the design review. It tends to be structural or invisible. Navigation organised around your internal departments rather than how buyers group information. Pricing or specification pages that make comparison difficult, so nobody can reach a confident decision. Error states nobody has looked at. Forms that ask for information the visitor has no reason to hand over yet.
That last point has evidence behind it. HubSpot's analysis of more than 40,000 landing pages found conversion declining as form fields were added, with more demanding inputs such as text areas and dropdowns associated with particularly pronounced reductions. The lesson is not that every form should have three fields. It is that every additional piece of information you demand needs to earn its place.
Useful questions here:
- Can a visitor who has never heard of you tell, within a minute, whether you serve businesses like theirs?
- Does your site acknowledge the alternatives a buyer is weighing up, including doing nothing?
- Do the pages your sales team actually rely on exist, and do they say what those conversations say?
- Where does the drop happen, and does that page have a defined job?
Problem three: the journey is fine and the measurement is wrong
Before spending anything, it is worth ruling out the possibility that the problem is the reporting.
The recurring versions of this are easy to check. The conversion event is defined at a point that does not correspond to commercial value, so form fills are counted and qualified opportunities are not. Sessions from non-buying audiences sit in the same average as buyers. Analytics and CRM numbers disagree and nobody has reconciled them, so both are quoted selectively. Consent and tracking gaps remove a chunk of the picture and the remaining data gets treated as complete.
This is not unusual. Ruler Analytics found in 2026 that 21% of marketers running paid activity struggle to track it effectively, rising to 25% for paid social, while a quarter use four or more tools to understand paid performance. The more systems involved in describing the same customer journey, the easier it is for apparent precision to conceal a fairly basic tracking and attribution problem.
The reason to fix this first is practical rather than pedantic: the other two diagnoses depend on being able to segment. You cannot tell a traffic quality problem from a journey problem if every visitor is in one bucket. When marketing and sales quote different numbers about the same pipeline, that disagreement is the first thing to resolve, and it is usually a marketing analytics and reporting job rather than a website one.
Five things worth doing before you approve more spend
None of this needs a large programme. Most of it can be done inside a fortnight with people you already have.
1. Segment before you judge anything. Split conversion by source and intent tier: branded search, non-branded search, paid, referral, returning visitors. Compare like with like. The moment you do this, one of the three problems usually becomes obvious.
2. Watch twenty sessions end to end. Specifically, sessions where somebody reached a decision-stage page and did not act. Aggregate data tells you where people stop. Recordings tell you why, and they tend to end internal arguments faster than a deck does.
3. Run a five-person clarity test. Show five people in your target role a key page for sixty seconds, then ask them what you do, who it is for, why they would choose you, and what happens next. If they cannot answer, that is a messaging problem and no redesign will resolve it.
4. Audit the promise against the page. Take your ten highest-spend search terms or campaigns and put the ad copy next to the page it lands on. Gaps between what the ad implies and what the page delivers are one of the most common and most fixable causes of expensive traffic that does nothing.
5. Reconcile one number with sales. Pick your primary conversion metric and check whether the sales team recognises it. If they do not, that is where to start.
RSPB Shop is the clearest example of what points one to four produce together. The work was structured conversion optimisation on an existing journey rather than a rebuild or a traffic programme, and it delivered £1.2m in incremental revenue at roughly 1,000% return on the investment.
The useful part is not the size of the numbers. It is what they demonstrate: the same traffic became worth more. Every pound already being spent on acquisition improved, without buying a single extra visitor.
When more traffic genuinely is the answer
Sometimes it is, and the argument in this article can be overapplied.
More traffic is the right investment when the journey already converts well for a clearly defined segment, you have not saturated that segment, and the unit economics still hold at the margin. If cost per enquiry is stable as you scale and sales are converting those enquiries, the constraint is reach rather than persuasion.
Buy more.
It is the wrong investment when the conversion rate for your best-fit segment is weak, when nobody can explain which visitors are worth having, or when the cost per enquiry has been climbing for two quarters and the response so far has been to increase the budget. If that last one sounds familiar, the problem may be further back still: what a healthy pipeline actually looks like covers the version of this that shows up after the enquiry arrives.
You probably do not need a new website yet
A rebuild is the most expensive answer available and it is frequently chosen before anyone has established what is wrong.
Rebuilding makes sense when the platform itself blocks improvement: you cannot test, cannot publish without a developer, cannot measure properly, or the site's structure actively fights the way buyers make decisions. In those cases, optimisation work will keep hitting the same ceiling and web design and build is the right conversation.
If none of that is true, the honest test is whether you have run out of things to improve on the site you have. Most teams have not. A structured audit of an underperforming journey routinely surfaces more worthwhile changes than a year of delivery capacity can absorb, and those changes can be shipped while the bigger decision is being made rather than instead of it.
Positioning work should come first when different parts of your own business describe the company differently, or when the sales conversation that wins deals bears little resemblance to the website. That is not a design problem, and a rebuild will simply reproduce the confusion in a nicer typeface.
The question to answer before the next budget conversation
Traffic buys attention. The journey decides what that attention is worth. Those are separate jobs with separate economics, and treating them as one number is what leads to spending more to get the same result.
So before the next increase is approved, one question is worth being able to answer with evidence rather than opinion:
For whom is this site already working, and what specifically is stopping the next person like them?
If the answer is that you are not sure, that is a diagnosis worth funding before a campaign is. More on how we approach conversion optimisation, or, if the issue is the wider digital picture rather than one journey, growing through digital.
Find out which of the three problems you have.
We will segment your traffic, watch the sessions and tell you whether the issue is who is arriving, what happens when they do, or how you are measuring it. If the answer is that you should buy more traffic, we will say so.
