Your store spends £40,000 per month on paid advertising. Traffic arrives reliably, campaigns are competently managed, yet the conversion rate has settled at 1.2% and every quarterly review shows the same pattern: higher spend, flatter returns. The next budget meeting has one real question to settle: should the business buy more traffic, or fix the store experience first?
The traffic or conversion decision shapes whether the next pound of budget compounds or evaporates. This guide gives owners, directors and managers of established stores a practical framework for making it: the numbers to check, the maths behind the trade-off, and the situations in which each investment tends to win.
The Short Answer: Fix Conversion First When the Store Underperforms Its Category
In most cases where a store spends heavily on advertising but converts below its category benchmark, conversion work should come first. A below-benchmark rate means the store is paying for demand it cannot capture, so each additional pound of traffic spend tends to be taxed by the same leaks. Scaling budgets on top of an underperforming funnel usually raises customer acquisition cost faster than it raises revenue.
This is not universal, however. Buying more traffic first can be the better investment when conversion already sits at or above the category norm, when traffic volume is too low to diagnose problems reliably, or when the business is entering a new market where visibility itself is the constraint. The wrong answer is the reflexive one, in either direction, taken without looking at the store’s own data. The rest of this article explains how to establish which situation applies to your store.
Traffic or Conversion: The Four Numbers That Make the Decision
The traffic or conversion question should not be settled by instinct, or by whichever agency argues its case most confidently. Four measurable checks decide it in most established stores.
1. Conversion Rate Against Category Benchmarks
A conversion rate only means something in context. Shopify’s benchmark analysis cites Statista data showing that 1.6% of global ecommerce visits converted into purchases in Q3 2025, with wide variation by category: luxury and jewellery sits near 0.94% while food and beverage reaches 6.22%.
A figure in the low single digits can therefore be below average for one category and perfectly healthy for another. Before allocating budget, compare your rate against what a good ecommerce conversion rate looks like for an established store. A store converting well below its category norm will usually get a better return from conversion work than from additional spend.
2. Whether the Funnel Data Can Locate the Leak
Conversion investment pays off fastest when the data already shows where buyers leave. Before deciding, the ecommerce team should be able to answer these questions:
- Where is drop-off largest: product page, cart, or checkout?
- Does mobile convert at less than half the desktop rate?
- Do paid landing pages convert worse than the site average?
- Is cart abandonment materially worse than the 70.22% average that the Baymard Institute has documented across 50 studies?
- Are revenue and conversion tracked reliably enough to compare month to month?
If these questions cannot be answered, fix measurement before making either investment. A store that cannot see where buyers leave risks spending blindly in both directions, and any later test results will be difficult to trust.
3. The Direction of the ROAS and CAC Trend
Return on ad spend that declines quarter after quarter, while creative and targeting remain competent and conversion stays flat, usually points to the store rather than the channel. In that pattern the advertising is delivering people, and the site is not turning enough of them into buyers.
By contrast, stable ROAS with audiences that are far from saturated suggests genuine traffic headroom. In that situation, scaling spend can be rational even before conversion work begins, because each additional pound still buys orders at an acceptable cost.
4. Whether There Is Enough Traffic to Fix Conversion
Conversion optimisation is an evidence-driven process, and evidence needs volume. A store with the level of ad spend described in the opening typically has tens of thousands of sessions per month, which is enough to diagnose behaviour and run meaningful experiments. A store with only a few thousand sessions often cannot, and structured A/B testing becomes slow or statistically unreliable.
For low-traffic stores, buying traffic first is often the correct sequence, because it creates the data that conversion work depends on. The budget question then becomes one of patience: the store buys visibility first, accepts a period of thinner returns while behavioural data accumulates, and schedules the conversion review for the point where the numbers can genuinely support one.
The CAC Maths on £40,000 per Month
The trade-off becomes concrete when it is expressed as cost per order. The figures below are illustrative rather than benchmarks: substitute your own click costs and order values.
Assume the monthly budget buys clicks at an average of £1.00, delivering roughly 40,000 paid visits. At the conversion rate from the opening scenario, those visits produce about 480 orders, which puts the paid customer acquisition cost at around £83 per order.
Now hold the spend constant and lift conversion to 1.6%, the global average in the Statista data above. The same visits produce about 640 orders, and acquisition cost falls to £62.50. That is a 25% reduction in CAC without a single extra pound of media spend. Buying those additional orders at the original rate would instead require raising the budget to roughly £53,000 per month, every month.
The asymmetry goes further. Media spend buys each visit once, while work to increase ecommerce conversion rate applies to every future visitor from every channel: paid, organic, email and direct. This is why conversion gains tend to compound while purchased traffic generally does not.
Investing in More Traffic vs Investing in Conversion: Direct Comparison
| More paid traffic | Conversion improvement | |
|---|---|---|
| Best suited for | Stores converting at or above category norms with unsaturated audiences | Stores converting below benchmarks with visible funnel drop-off |
| Time to first impact | Days to weeks | Weeks to months, including diagnosis and testing cycles |
| Cost structure | Ongoing media spend that stops producing when paused | Project or retainer fees that produce lasting changes |
| Effect on CAC | Usually raises CAC as spend scales into colder audiences | Lowers CAC across every channel, not only paid |
| Compounding | None; each visit is bought once | Yes; each improvement benefits all future traffic |
| Main risk | Scaling budget into an unfixed funnel | Slow or poorly run testing that consumes months without results |
| When it fails | Audience saturation, rising click costs, weak store experience | Insufficient traffic volume or no implementation capacity |
Read this way, the comparison is less a contest than a sequencing guide. Paid traffic remains essential for growth and acts within days; conversion work starts more slowly, but its gains persist and apply to every channel. Fixing conversion first tends to make each later pound of traffic spend more productive, provided there is a measured problem to fix rather than an assumed one.
When Buying More Traffic First Is the Right Call
A balanced decision framework has to acknowledge the situations where additional traffic is the better first investment:
- Conversion is already competitive. A store at or above its category norm has less friction to remove, and further conversion gains tend to arrive slowly and expensively.
- Traffic is too low to diagnose. Without enough sessions, behavioural data is mostly noise and experiments cannot reach reliable conclusions.
- The business is entering a new market. Awareness, not friction, is the constraint when few people in the target market know the store exists.
- A seasonal window is open. Demand peaks reward visibility immediately, while conversion projects rarely complete inside a single season.
- Retargeting pools are too small. Prospecting spend builds the audiences that later campaigns and email flows depend on.
Even in these cases, traffic-first is a sequencing choice rather than a permanent strategy. Once volume and data exist, the conversion question returns, and it tends to return with sharper teeth: the more a store spends on acquisition, the more each remaining point of friction costs in absolute terms.
How to Run the Decision in Practice
- Validate tracking. Confirm that orders, revenue and channel attribution are recorded accurately. The rest of the process relies on this being trustworthy.
- Segment the conversion rate. Split it by device, traffic source and new versus returning customers. A blended average often hides a far worse figure for paid prospecting traffic on mobile.
- Benchmark against the category. Establish whether the store underperforms comparable retailers or simply reflects its product type.
- Map the funnel drop-off. Quantify losses from landing page to product page, cart, checkout and payment, and identify the single largest leak.
- Model the CAC trade-off. Using your own click costs, calculate what a realistic conversion improvement would do to cost per order versus the same money spent on media.
- Allocate and sequence. Hold advertising at its efficient level, and direct incremental budget to fixing the largest measured leak first.
- Reassess quarterly. As conversion improves, the marginal return on traffic rises, and the right allocation shifts back towards acquisition.
This sequencing is visible in WD Market’s own project work. In the Profcentrs case study, two brands were unified into one store with conversion-driven infrastructure improvements, and the published results report a returning customer rate increase of 71%. Repeat purchases of that kind arrive without additional media spend, which makes the acquisition budget already in place work harder rather than demanding more of it.
Common Mistakes When Paid Ads Stop Paying Off
None of the following mistakes indicates a careless team. Each is a reasonable-looking response to commercial pressure that tends to fail in a predictable way, which is precisely what makes them common.
The most expensive mistake is scaling spend to outrun the problem. When returns flatten, adding budget can feel decisive, yet the same funnel leaks drain the larger amount and acquisition efficiency usually falls further. The quarterly review then repeats itself, one budget increment later and with less room to manoeuvre.
The opposite reaction, pausing all advertising to focus on conversion, tends to backfire as well. Modern bidding systems learn from conversion volume; Google’s Smart Bidding optimises for conversions using auction-time data. Cutting spend to zero starves those algorithms and removes the very traffic that behavioural analysis and testing depend on, while retargeting pools shrink at the same time, which can make restarting slower and costlier later.
Judging the store on a blended average is a subtler error with similar consequences. Brand search and returning customers usually convert well and lift the overall figure, which can hide how poorly cold paid traffic performs. A decision made on the blended number often solves the wrong problem, because the segment consuming the media budget is not the segment holding the average up.
Starting with a redesign is another frequent detour. Rebuilding the store before diagnosing where and why buyers leave tends to move the same friction into a new theme at significant cost, and it resets whatever behavioural data existed. A redesign can be the right outcome of a diagnosis; it is rarely a substitute for one.
Finally, buying analysis without implementation capacity changes little. An audit that the development team has no time to build sits in a drawer while acquisition costs stay where they were. Before commissioning conversion work, it is worth confirming who will ship the changes and on what schedule.
Key takeaway: the traffic or conversion question is a sequencing decision rather than a loyalty test. Fix the store first when it converts below its category norm and the funnel shows where revenue leaks; buy traffic first when conversion is already competitive and the audience still has room to grow.
Conclusion
A store that spends heavily on advertising while converting below its category benchmark will usually gain more from fixing the buying experience than from buying additional visitors, and the reverse holds when conversion is already competitive and audiences still have headroom. Check the rate against category norms, confirm the funnel data can locate the leak, read the ROAS trend honestly, and run the acquisition-cost arithmetic with your own numbers. The decision then stops being a debate between vendors and becomes a calculation the business can defend, revisit each quarter, and adjust as the store itself improves. That discipline, more than any single allocation, is what separates budgets that compound from budgets that merely repeat.
Frequently Asked Questions
Should we pause paid advertising while fixing conversion?
Usually it is better to trim than to switch off. Holding a reduced but efficient core of spend keeps sessions and purchase events flowing, which both the diagnosis and any split tests rely on, and it prevents remarketing audiences from emptying while the work happens. Stores that stop completely often find that rebuilding campaign momentum takes longer and costs more than expected.
Is a 1.2% conversion rate a problem for an established store?
That depends on what the store sells. For most product categories it sits below the typical range, and paired with heavy ad spend it usually signals capture problems worth diagnosing. For high-consideration purchases such as fine jewellery, however, rates below 1% are common, and treating a normal figure for the category as a failure can push budget in the wrong direction.
How much budget should move from advertising to conversion work?
There is no universal split. A practical method is to find the least efficient slice of current spend, often the final scaling increment where returns are weakest, and redirect that amount into diagnosis, testing and implementation. At the spend levels discussed in this article, even a modest reallocation can fund a serious conversion programme without a visible dent in traffic volume.
How quickly does conversion work reduce customer acquisition cost?
Rarely in the first month. Diagnosis normally takes a few weeks before anything ships, and each split test needs enough conversions to give a trustworthy read, typically two to four weeks. The effect on acquisition cost builds as winning changes accumulate, with fixes to measurable technical friction, such as a slow mobile checkout, tending to land first.
Can we invest in traffic and conversion at the same time?
Yes, and established stores generally end up doing both continuously. The traffic or conversion framing concerns where the next incremental pound goes, not a choice of allegiance. Advertising continues at whatever level remains efficient, conversion work draws its funding from the weakest spend, and the balance drifts back towards acquisition as the store converts more of what it already buys.
What if the store does not have enough traffic to A/B test?
Larger, bolder changes can show an effect with fewer conversions than small tweaks, and carefully monitored before-and-after measurement can stand in for split testing when volume is thin. Building traffic first is also a legitimate answer here, since it creates the visitor volume that structured experimentation later needs.
Deciding Where the Next Pound Goes
If the traffic or conversion checks in this article point at the store, the productive next step is a structured diagnosis rather than a redesign or another budget increase. Request a CRO audit from WD Market to receive a prioritised list of the conversion problems currently taxing your advertising spend, with a clear recommendation on what to fix first. For ongoing analysis of decisions like this one, follow WD Market’s ecommerce insights on LinkedIn.