With the same ad budget,
less traffic: twice the orders,
2.6× the conversion rate.
We kept ad spend steady and rebuilt the on-site experience. The difference came from conversion, not budget.
For client confidentiality, absolute figures are indexed or presented qualitatively; rates, percentages, and multipliers are unchanged.
The problem was not traffic.
A founder-led, design-driven DTC brand in a category with a high average order value and low purchase frequency; traffic was primarily mobile and social.
Only 1–2 out of every 1,000 visitors was buying. As advertising costs rose rapidly (cost per click increased approximately 2.9× in one year), the right move was not to spend more — it was to convert more of the visitors already arriving.
We held the advertising variable steady.
To show that a conversion improvement truly came from the site work, we had to control for advertising. We therefore built the comparison around two months in which Meta spend was almost identical.
| Metric | Before | After | Change |
|---|---|---|---|
| Meta ad spend | 100 | 98 | ≈ same |
| Visitors (sessions) | 100 | 78 | −22% |
| Orders | 100 | 210 | 2.1× |
| Conversion rate | 0.14% | 0.37% | 2.6× |
| Revenue | 100 | 212 | 2.1× |
| Meta ROAS | 3,04 | 5,35 | 1.8× |
Indexed results with the same budget
Before = 100 · Absolute figures are indexed for confidentiality; rates and multipliers are real.
Absolute figures are indexed for client confidentiality (before = 100). Rates and multipliers are real data.
Advertising cannot explain this difference.
Every step in the funnel
improved on its own.
Looking back one year:
orders increased 2.8×.
Orders increased 2.8× while visitor volume stayed almost the same.
Starting point
Before the work
Starting point + 3
Month 3 of CRO
Starting point + 6
Month 6 of CRO
Conversion rate · single source · non-campaign months
| Pre-work period | Advertising status | Conversion |
|---|---|---|
| Reference month | Active | 0.13% |
| Second month | Active — 3× the reference month budget | 0.11% |
| Third month | Active — 2.6× the reference month budget | 0.07% |
| Ad-free month | Almost no advertising | 0.32% |
Why the ad-free month is separate: There was almost no advertising that month. Visitors came entirely through the brand’s own traffic: social, search, and direct visitors who already knew the brand. The rate was 0.32% under those conditions; when ad traffic was introduced, it fell to 0.07–0.13%.
The improvement was gradual
and sustained.
This was not one unusually good month, but a steady climb. Below is the monthly conversion rate for the full period — including two abnormal months, clearly marked.
2025
2026
bot traffic
2026
sale
2026
2026
2026
2026
2026
The trend is clear in the remaining non-campaign months: 0.13% → 0.14% → 0.24% → 0.22% → 0.31% → 0.37%. When advertising nearly stopped for one month, the smaller volume of warm traffic naturally lifted the rate — the same effect was visible in the ad-free month the year before (0.32%). The real signal is this: when advertising returned at full budget, the rate did not fall; it reached the period high of 0.37%.
The funnel did not leak when scale arrived.
- Rebuilt mobile purchase flow
- Product-page improvements
- Trust-building elements at checkout
- Cross-sell structure in the cart
- Stronger social proof
- Site-wide user experience improvements
- Simplified signup and contact flows
- Automated email flows (abandoned cart, post-purchase)
When the brand concentrated its ad budget on a single market, traffic doubled in one month. Under that kind of load, conversion usually falls — because the incoming traffic is cold. Here, it did not fall; it rose.
The pre-built funnel absorbed the surge without losses, and return on ad spend (ROAS) reached 5.35. That is the clearest practical benefit of CRO: making it safe to scale the advertising budget.
The abandoned-cart automation we built also recovered directly attributable additional sales over a seven-month period.
Strengthen the funnel
before scaling the ad budget.
A 30-minute analysis session for founder-led, design-driven premium DTC brands that are mobile- and Instagram-heavy. We will review your own dashboards together.
Let’s find where your data is failing to become sales.
Method: Visitor, cart, checkout, and order data came from a single source (Shopify Analytics); spend, ROAS, and CPC data came from Meta Ads Manager. Figures from different systems were not mixed into a single rate; Meta figures use Meta attribution (7-day click). Because monthly order volume is low, we assessed the period trend rather than individual months.
info@ajanweb.com ↗