CPC & CTR Benchmarks: Italian Data, Global Method

“My agency says €1.80 a click is normal for my industry. Is that true?” Plenty of small business owners ask this the first time they open a Google Ads report. The answer usually comes back with a number lifted from an American study.

Right now, there’s no public, up-to-date, industry-by-industry Google Ads benchmark for Italy. In our research we found no official Google averages broken down by country. What you can do is build yourself a reference point that beats any national average. You need two things: what your own keywords actually cost, pulled from Google’s tools with the location set to Italy, and a way to read CTR and conversion rate without lumping together campaigns that measure different things.

This page covers the Google Ads side of digital advertising benchmarks in Italy, the wider picture that brings together Google, Meta and the other platforms. We stick to Google Ads almost the whole way through. Other platforms only show up where a comparison helps.

Is there an official Google Ads benchmark for Italy?

No. We found no official Google averages for CPC, CTR or conversion rate by industry and by country. For Italy you have two reliable data sources. The first is Keyword Planner, set to location Italy and language Italian. The second is your own account history.

We’re putting this up front because it changes how you read everything else. If a slide shows you the “average CPC in Italy” for an industry with no year, source or currency attached, that number is worth about as much as a guess over coffee.

For this article we drew on three kinds of source:

• academic research on search advertising, which explains how clicks and conversions behave;
• documentation of the platform and its ad networks, which shows why some campaigns can’t be compared;
• a data pull you can repeat yourself in an afternoon with Keyword Planner. We walk through it step by step below.

Before you compare any number, watch out for three traps.

1. Currency. A CPC in dollars, measured in American auctions, is still American data after you convert it to euros. The exchange rate changes the unit. The market stays the same.
2. What counts as a conversion. A 10% rate based on clicks on your phone number and a 2% rate based on signed quotes are measuring two different things.
3. The brand vs non-brand mix. A campaign that also bids on your company name will have a CTR inflated by people who were already looking for you.

The same approach applies to B2B ad benchmarks, where the lack of Italian data hurts even more.

Average Google Ads CPC in Italy on the Search and Display networks

What the international data says

The annual reports with industry tables that we found mostly cover the US market, with US budgets, US competitors and US prices, on a completely different scale. A review of deep learning approaches to CTR prediction notes that online advertising revenue in the United States grew 12.2% in 2020.¹ In an auction that size, what “divorce lawyer” costs in Chicago tells you almost nothing about what the same click will cost in Verona.

That’s why we’re not reproducing the dollar tables here. We couldn’t verify them against the original reports. Even if we had, they’d have pointed you at the wrong yardstick.

What’s worth taking from the international literature is a concept, and it’s more useful than any figure. A survey of CTR prediction models published by Cambridge University Press sums up an ad platform’s expected revenue as R = CTR × CPC.² For Google, an ad that gets clicked a lot earns more every time it’s shown. So read CTR and CPC together, in the same table.

Building an Italian benchmark with Keyword Planner

Hardly anyone explains this step in Italian, and it’s worth more than any published average. You need a Google Ads account (no live campaigns required), a Google Sheet and about two hours.

There are five steps.

1. Open Keyword Planner and choose the option for search volume and forecasts.
2. Set the location to Italy, or to your province if you sell locally. Set the language to Italian and the network to Google Search.
3. Enter 10-15 commercial keywords: searches made by people who want to buy or ask for a quote. Leave informational ones out.
4. For each keyword, record average monthly volume, the two top-of-page bid values (low and high), the competition level and the date you pulled the data.
5. Repeat the pull every quarter on the same list. After a year you’ll have a time series no public study can give you.

Here’s a starting template you can adapt to your industry.

Industry	Starting keyword	Why it’s a good reference
Business services	commercialista milano	Contact intent, heavy local competition
Private healthcare	dentista impianti prezzo	People searching for price are close to deciding
Legal	avvocato divorzio + city	High value per client, expensive auctions
Real estate	agenzia immobiliare + city	Shows how much location drives price
Software	software gestionale pmi	B2B search with a long sales cycle
Home services	idraulico pronto intervento + city	Urgent, lots of calls from mobile
Furniture	cucine su misura + city	Big ticket, slow decision
Training	corso excel aziendale	B2B demand with modest budgets
Catering and events	catering matrimonio + province	Highly seasonal, pull it across several months
Automotive	noleggio lungo termine partita iva	Strongly commercial, national competition

Treat the two bid values as a starting range. Once you have two to four weeks of data, compare that range with the CPC in your account. The gap between the two already tells you something. If you’re consistently paying above the high value, the problem is usually the ad or the landing page, and rarely the industry.

Take a dental practice in Brescia trying to work out whether €1.80 a click on “impianti dentali” (dental implants) is expensive. With fourteen keywords pulled for the province, the owner sees what their own searches cost in their own area, and stops relying on a generic average.

How AI can speed this up. Export the CSV from Keyword Planner and hand the suggested keywords to Claude or ChatGPT to sort them by intent before you pick your 10-15. This prompt works well.

“Below is a keyword list exported from Google Ads Keyword Planner, with monthly volume and top-of-page bid. My business is [one-line description] and I operate in [area]. Put each keyword into one of four categories: commercial (wants to buy or get a quote), informational (wants to understand something), brand (contains a company name) or off-target. Return a table with keyword, category, volume and a reason in ten words or fewer. Then list the 15 commercial keywords with the best ratio of volume to high bid, excluding searches outside my area.”

Check the classification by hand. The model slips up mostly on ambiguous searches like “dentista economico” (cheap dentist), which can come from someone ready to buy or someone just curious.

Search and Display can’t be compared

Google Ads shows ads in very different places. On search results pages, up to four ads appear above the organic results, mid-page or at the bottom. The same account also buys space on YouTube, Maps, Discover and third-party sites and apps, which together make up the Display Network.³

On Search you pay for a click from someone who has just typed what they want. On Display you’re interrupting someone who was reading something else. So a click means something different. A team of researchers at NYU Stern analysed 58 display campaigns and found that clicks were poor predictors of purchases, in many cases no better than random. Visits to the brand’s website turned out to be a far more reliable signal.⁴

Averaging Search CPC with Display CPC gives you a meaningless number. Keep them on two separate rows of your report, always.

Average Google Ads CTR changes with campaign type

CTR is clicks divided by ad impressions.² The maths is simple. The result depends almost entirely on where and to whom you show the ad.

A brand campaign, bidding on your company name, has a sky-high CTR because someone searching “Rossi serramenti Padova” is looking for you. A non-brand campaign on “serramenti pvc prezzi” (PVC window prices) is up against ten competitors and will have a much lower CTR. Shopping, Display, Performance Max and Demand Gen each work by their own logic. If your report shows a single “average account CTR”, ask for it broken down by campaign type and by brand vs non-brand. Without that breakdown, the number tells you nothing.

Then there’s position. Chongyu Lu (Pace University) and Rex Yuxing Du (University of Houston) studied Google AdWords data from 13 advertisers. They found that users favour the first paid ad, and that click behaviour shifts as an ad drops down the page, with differences between smartphone, desktop and tablet.⁵ A low CTR on mobile might come down to a bid that keeps you in second or third place, even when the copy is fine.

Copy still matters, in a way that’s not obvious. A study from Tilburg and Maastricht published in Marketing Letters compared different types of argument in search ads. Ads quoting an expert or citing a statistic got more clicks. Ads explaining cause and effect (“cut your energy bills because the glass is triple-glazed”) got more conversions.⁶ Your best-CTR ad and your best contract-winning ad may well be two different ads. Working on copy helps lift CTR, and a higher CTR is only worth something if contracts follow.

Average CPC by industry also varies from platform to platform, and Google’s costs should be read next to those of your other channels.

How AI can speed this up. The search terms report is where you start splitting brand from non-brand. Export it and use this prompt.

“This is the search terms report for my Google Ads campaign over the last 90 days (columns: term, impressions, clicks, cost, conversions). My brand is [name] and people write it as [variants, including misspellings]. Split the terms into brand and non-brand. For each group, calculate CTR, average CPC and cost per conversion. Then list the 20 non-brand terms with the highest spend and zero conversions.”

That gives you two CTRs and two CPCs, each one comparable over time.

Conversion rate depends on what you count as a conversion

Two small businesses in the same industry can show conversion rates of 12% and 3% and be running equally well. All it takes is for the first to count a tap on the phone number as a conversion while the second counts only completed quote forms. So treat any cross-company conversion rate comparison with real caution if it doesn’t state which definition it uses.

The NYU Stern research mentioned above applies here too. Optimising for an easy-to-collect signal like the click pushes campaigns towards people who click, at the expense of people who buy.⁴ The same thing happens when an account optimises for a micro-conversion, such as a visit to the contact page. The algorithm finds loads of people happy to visit your contact page, and few of them become customers.

Get the order right in four steps.

1. Write down your conversion hierarchy, from the action closest to revenue (contract, order, accepted quote) to the furthest (phone tap, download).
2. Set only the actions at the top as primary, meaning they’re used for bidding. The rest stay visible as secondary.
3. Calculate conversion rate on primary actions only, and state in the report which definition you’re using.
4. Once a month, match conversions against your CRM to see how many turned into customers.

In e-commerce the conversion is the purchase, so comparisons are cleaner. That’s why e-commerce ad benchmarks focus mainly on order value and ROAS. In services, where weeks can pass between click and contract, settle the definition before you launch the campaign.

How AI can speed this up. Before you even open the account, use Claude or ChatGPT to build the hierarchy from your sales process.

“I sell [service] to [type of customer]. The typical path is: [describe the steps, e.g. Google search, phone call, site visit, quote, signature]. My average contract value is [amount] and I close roughly [x] quotes out of 10. Propose a hierarchy of Google Ads conversion actions, from closest to revenue to furthest. For each, say whether it should be primary or secondary and how I’d track it (form, call, CRM import). Flag any actions likely to inflate the conversion rate without bringing in customers.”

What to measure instead of CTR on Performance Max and Demand Gen

For these campaigns, CTR and CPC figures get passed from blog to blog with no method and no source. Don’t use them.

The problem is structural. When a campaign spreads ads across very different placements, its CTR averages impressions that have little in common. The figure moves every time the algorithm shifts budget between placements, even if the campaign’s quality hasn’t changed at all. Comparing it with a Search campaign’s CTR makes no sense, and comparing it with another account’s average makes even less.

On Performance Max and Demand Gen, watch these four indicators.

• cost per primary conversion, using the strict definition you set in the previous section;
• the share of conversions coming from searches for your brand, because a Performance Max campaign living off brand traffic is buying customers you already had;
• conversion value against spend, if you run an online shop or import contract values from your CRM;
• a test with a control area: a province where you switch the campaign off for a few weeks to see how far the total drops.

That last one is the most work, and it’s the only one that measures what the campaign really contributes. How these automation-driven campaigns evolve is one of the AI advertising trends for 2026.

Treat Quality Score as a diagnostic tool

In plenty of agency reports, Quality Score shows up as a grade to push higher, with the promise that a higher score means a lower CPC. You’ll get more out of it as a warning light that points you, keyword by keyword, to where you should look. Keep it out of the monthly summary.

In your account, add the Quality Score columns at keyword level and sort by spend. For your top twenty keywords by cost, check which component is flagged as below average. Each one calls for a different fix. If it’s expected CTR, work on the ad copy. If it’s ad relevance, the keyword is sitting in the wrong ad group. If it’s landing page experience, the work is on the website.

That last case comes up a lot. The campaign sends traffic to the home page, the home page talks about ten services, and someone who searched “cucine su misura Treviso” (made-to-measure kitchens in Treviso) can’t find the kitchens straight away and leaves. Tweaking bids won’t do much here. What fixes it is a page built for that search, with the service, the area and a quick way to ask for a quote.

My CTR is below average. Now what?

Before you change anything, ask whether you’re comparing yourself to the right average. Here’s an order of work that holds up.

1. Compare against your own history. Same campaign type, same months last year, same conversion definition.
2. Segment. Brand vs non-brand, Search vs Display, mobile vs desktop. A lot of “drops” vanish once you separate the campaigns.
3. Look at primary conversions. A below-average CTR with a sustainable cost per contract can be left alone.
4. Work out the highest CPC you can afford. That’s the benchmark that really counts.

Here’s an example for that last step, with numbers you’d swap for your own. A window and door company makes €1,200 margin per customer and closes one quote in five. To break even, it can spend up to €240 per quote request (1,200 divided by 5). If the landing page turns 5% of visitors into requests, the maximum sustainable CPC is €12 (240 × 0.05). On that maths, €1.80 a click sits well under this company’s ceiling, and whether it’s expensive or cheap depends only on that ceiling. For another business with thinner margins and a page converting at 1%, the same €1.80 would already be at the limit.

Run the same numbers when you’re weighing up another channel. Comparing Google Ads with Meta Ads, or looking at Meta Ads benchmarks for Italy, only becomes useful once you know your spending ceiling per customer.

How AI can speed this up. When the monthly report lands from your agency or freelancer, give it to Claude with this prompt before the meeting.

“Attached is this month’s Google Ads report. My average margin per customer is [amount], I close [x] requests out of 10, and I count only [definition] as a primary conversion. 1) Check whether the report separates brand from non-brand and Search from Display, and whether it states the conversion definition. 2) Calculate the maximum sustainable CPC from my numbers and compare it with the actual CPC per campaign. 3) Write five specific questions to put to the agency wherever the report doesn’t let me tell whether the spend is sustainable. No generic commentary on performance.”

Those questions shift the meeting from “CTR went up” to “what did each customer cost us”. That second number is the one you can actually make decisions on.

Fonti

Footnotes

1. A Review of Click-Through Rate Prediction Using Deep Learning. mdpi.com ↩

2. A comprehensive survey on advertising click-through rate prediction algorithm. cambridge.org ↩ ↩²

3. Google Ads. it.wikipedia.org ↩

4. Evaluating and Optimizing Online Advertising: Forget the Click, but There Are Good Proxies. archive.nyu.edu ↩ ↩²

5. Click-Through Behavior across Devices In Paid Search Advertising. doi.org ↩

6. Search engine advertisements: The impact of advertising statements on click-through and conversion rates. doi.org ↩