Website
More quality inquiries
A clearer offer, a better path through the site, and calls to action that help a visitor take the next step.
Our wins
Results you can actually read and improve on.
We do not pick only the numbers that look good. For every project we first decide which change actually has value for the business.
What we measured
Six results from the reports our clients received
Each one comes from a report the client received by email. For each one we also say what was wrong with it, or what we have not solved yet.
- +172%orders per dayTartuf.com, after the ad went live
- +131%revenue per dayTartuf.com, the first eight days
- €18.24of revenue per 1 EUR of ad spendGhost.si, May 2026
- +72%inquiries in the Italian marketSimak Keramika, in one month
- +50%inquiries in one monthIN2050, Google Ads
- −24%cost per inquirySimak Keramika, in one month
Case · Google Ads
IN2050: 50 percent more inquiries in one month
The company sells across seven European markets and in the US. The campaigns were running, inquiries were few, and nobody could say how many there really were.
What we found
The first finding was not about the ads but about the measurement. Google Ads was recording zero conversions, even though the forms on the site worked. There were three causes, each enough on its own: the tag fired on an event nobody ever triggered; the tracking looked for form markers that exist on only one page, not on the home page or the translated ones; and one form returned an error on every submission because the recipient was configured wrongly.
In July, two thirds of submitted forms were never recorded. The system saw a third of the truth and steered the budget by it.
What we did
We fixed all three causes, built a single event that fires on every successful submission, and connected it to Google Ads and Analytics 4. Alongside that, month after month, we cleaned the search terms, tidied the keyword structure and moved budget toward the markets producing inquiries at a lower cost.
What came out of it
From June to July 2026 the number of genuine inquiries rose by 50 percent, and the cost per inquiry fell by 11 percent. Of all the submissions in July, 73 percent were genuine inquiries, once duplicates and one irrelevant message were subtracted. We wrote that to the client too, because the difference between a submission and an inquiry is what the budget is judged on.
The click-through rate was 9.72 percent. For search advertising that is high, and it means the ads answered the same question the person typed.
What is not solved yet
The site still has no consent platform, which has to be sorted out for campaigns aimed at the EU. Once it is, there will be fewer measured conversions, not more, because some are currently recorded without consent. We told the client that in advance, not after the numbers dropped.
The numbers come from the monthly reports the client receives every month. You get the same report, including in a month that was not good.
Campaign running and no inquiries? The first thing we check is whether you are measuring the right thing at all.
Request a campaign review →Case · three channels, three months
Simak Keramika: a quarter lower cost per inquiry
A tile retailer with a showroom and an Italian market. We run Google Ads, Meta ads and email at the same time. Below are three consecutive months, not one cherry-picked.
Google Ads, May to July 2026
Clicks went 1,263, then 1,499, then 1,582, so 25 percent more in three months. Impressions went 12,018, then 14,214, then 16,086. The cost per click stayed at 0.20 to 0.21 EUR throughout, and spend between 248 and 320 EUR a month. Traffic grew; what it cost did not.
The location campaign for the showrooms is the most telling figure: the click-through rate rose from 28.23 to 29.84 and then to 31.57 percent. One in three people who saw the ad clicked it. Cost per click, 0.17 EUR.
Meta ads
In June the advertising brought 148 new contacts at an average of 2.98 EUR. In July, 138 contacts at 2.27 EUR, so almost the same volume on 28.8 percent less spend, which is roughly 127 EUR saved.
The Italian market moved most: from 39 to 67 inquiries, with the cost dropping from 3.81 to 2.32 EUR. The Slovenian campaign produced 71 inquiries at 2.23 EUR in July.
In May, deliverability was above 98.5 percent with not one spam complaint; in July, 99.89 percent. From seven campaigns in June to ten in July. Clicks in the Slovenian segment were 38 percent higher in July than in June, and opens in the Italian segment rose from 11.20 to 13.02 percent.
What three months show
None of these numbers is a jump. Each is a shift of a few percent in the right direction, every month. That is what work somebody is actually doing looks like, as opposed to a campaign somebody switched on and left running.
Three channels only make sense once each one knows its job, and once somebody looks every month at what moved.
Let us look at what you already have →Case · online store
Tartuf.com: 131 percent more revenue per day
A store selling fresh truffles and products made from them. The ad went live on May 24, 2026, so the difference can be measured directly: before that day and after it.
The first eight days
Daily revenue rose from 48.70 to 112.49 EUR, which is 131 percent more. Orders per day rose from 0.78 to 2.13, which is 172 percent more. Over eight days: 899.90 EUR of revenue and 17 orders.
Of that, Facebook and Instagram contributed 696.77 EUR and 12 orders, so 77.4 percent of all revenue after the ad went live. The ad spent 76.61 EUR, which is 6.38 EUR per purchase and roughly nine euros of revenue for every euro in.
June and July
In June the ad spent 301.18 EUR and produced 28 purchases worth about 1,662.51 EUR. Total store sales were 3,086.10 EUR, which is 52.8 percent more than May, with 25.7 percent more orders.
In July: 309.19 EUR of spend, 30 purchases worth about 1,521.21 EUR, at 10.31 EUR per purchase. Facebook’s share of the store’s total sales rose to 61.7 percent, and the average order value to 78.47 EUR.
Why this works at ten euros a day
The budget was ten euros a day throughout. At that level the amount is not what decides it; what decides it is whether the ad hits the right product and the right person, and whether the path to purchase is short enough. So we worked on the products and the buying journey rather than on raising the budget.
With the report we also proposed connecting the accounting software to the store, so invoices would no longer have to be issued by hand. That was not upselling a service; it was an hour a week that was being lost.
A small budget is not the obstacle. The obstacle is an ad pointing at a page that does not sell.
Let us look at your audience →Case · honesty before a nice number
Ghost.si: 18.24 EUR of sales for every euro spent
The Ghost bike store. In May 2026 the advertising produced the best ratio we have ever measured. In June we wrote in the report that the figure must not be read as sales, and why.
May 2026
On 257.79 EUR of spend, the store’s own data shows five completed orders worth 4,702.67 EUR in total. That is 18.24 EUR of completed sales for every euro spent.
A further seven orders worth 7,878.56 EUR were placed but never completed, because of installment financing through Slovenian providers (Leanpay and NLB Buy&Go). Total purchase intent was therefore 12 orders worth 12,581.23 EUR. The buyers reached the order stage; what stopped them was the financing, not the ad and not the store.
June 2026, and why we did not write the bigger number
In June the campaign produced 1,866 landing page views, 28 add-to-carts and 10 orders from Facebook and Instagram worth 36,584.92 EUR.
That figure we did not declare as sales. There can be a technical mismatch between the installment payment system and the store, so there was no way to know how many orders were actually approved. They were recorded in the report as purchase intent, and the next step was checking how orders were being completed, not raising the budget.
Why we put this on our own site
36,584 euros is a nicer number than 4,702. An agency that writes the first one without a footnote gets a better-looking report, and the client gets a false picture and a wrong budget decision behind it. We report this way even when it costs us.
If the measurement is not right, every number in the report is only an appearance.
Let us look at your tracking →What a good result means
The right metric depends on the right goal.
Different projects need different measures. Below are the areas we most often track and improve on.
Online store
Fewer obstacles to buying
Readable categories, useful product pages and a simple path from the cart to a completed purchase.
SEO and marketing
Visibility that is relevant
Presence in the searches and audiences connected to what you sell, and measurement of what they contribute to the goal.
AI automation
Less repetitive work
Shorter response times, less retyping by hand, and a better view of the process, the data and the exceptions.
Measurement without flattery
A number is only useful once it leads to a decision.
Before starting we set the baseline, the key events and how we will report. That way we can tell nice numbers apart from changes that mean something for the business.
We set the baseline
We review where things stand, so later changes are not explained without context.
We choose the key measures
We track a handful of metrics tied directly to the goal of the project.
We agree the next move
Every report has to explain what we learned and what makes sense to do next.
The next step
Numbers like these start with one intro call
All four cases above began with somebody telling us where they were and what was bothering them. Fifteen minutes, no preparation.
We also say when it is not worth it. For one client we wrote in the report that after six days the campaigns did not have enough data to judge. We would report the same way to you.
- You receive an itemized proposal the same or the next day
- The same monthly report you see in the cases above
- If we judge we cannot help you, we say so immediately
Prefer to call? +386 70 743 718, WhatsApp or info@aithena.co