Why do French and English ads have such different conversion rates? How can you identify the cause?

When conversion rates differ significantly between French and English ads, do not rush to rewrite the copy when investigating the cause. First confirm that both ad groups are measuring the same type of conversion and that their traffic conditions are comparable. Then check each step in the journey: ad click → page visit → inquiry or checkout initiated → conversion completed. The point at which the gap begins to widen usually helps narrow down the cause.

Also review the cost per qualified inquiry or completed sale. If English ads generate more form submissions but also more unqualified inquiries, looking at conversion rate alone could shift the budget toward lower-quality traffic.

Rule out measurement differences and confirm that conversions have actually declined

Conversion rates must use consistent numerators and denominators. Ad platforms often measure ad interactions, while website analytics tools may measure sessions. The two cannot be compared directly. If the French group counts successful form submissions but the English group also counts clicks on the submit button, the gap may be due to tracking settings.

Align the conversion events, reporting periods, attribution windows, and rules for counting repeat conversions across both ad groups, and confirm that they use the same optimization goal. B2B websites can distinguish between “form submissions” and “qualified inquiries,” while online stores should distinguish between “add to cart” and “successful payment.” Do not compare actions from different stages as if they were the same.

Next, complete the full user journey once on both the French and English pages. Check whether events fire and whether the backend receives the form submission or order. Pay particular attention to whether French form submissions fail because of field validation, whether the thank-you page is missing tracking code, and whether referral information is lost during cross-domain checkout.

If the two groups receive traffic from different regions, also check for differences in user consent status and tracking coverage. If the number of orders received by the backend has not declined but the ad platform reports significantly fewer conversions, address the measurement issue first. B2B inquiries and sales also take time to materialize, so do not directly compare a newly launched French campaign with an English campaign that has already accumulated a full set of conversions.

Separate language differences from traffic differences

French and English ads often target different countries, devices, and audiences at the same time. The results therefore reflect a combination of factors and cannot be attributed to language alone. For example, if the English group includes branded searches and remarketing while the French group mainly reaches people encountering the brand for the first time, it is not surprising that the latter has a lower conversion rate.

After exporting the data, first break it down by the following dimensions:

• Countries and regions: Review markets such as France and French-speaking Canada separately rather than grouping them all into one “French market.”
• Channels and traffic sources: Compare search ads, social media ads, branded keywords, non-branded keywords, and remarketing separately.
• Devices and pages: Check whether the gap is concentrated on mobile devices or on a particular product landing page.
• Search terms and audiences: Check for traffic related to tutorials, recruitment, free resources, or other searches that do not align with your business goals.

If the overall gap is large but narrows significantly when comparing the same regions, devices, and traffic types, adjust the traffic mix first. Ad language settings do not necessarily reflect users’ native languages, so also consider search terms and the pages users actually visit.

Compare click volume and conversion count as well. A small change in the number of conversions can cause a large fluctuation in the rate, so avoid stopping campaigns in a particular language based on short-term data. Also, do not segment the data too finely, or there may not be enough data in each group to support a reliable assessment.

Follow the conversion path to find where users first drop off

Build a simple funnel for both languages: ad clicks, landing page visits, key actions, and successful submissions or payments. Compare the number of users and the progression rate at each step, using consistent event definitions throughout.

Users do not reach the page properly after clicking: Check page load speed, link redirects, automatic language switching, and mobile compatibility. Opening an English page from a French ad, or repeatedly redirecting users based on their browser language, can disrupt the browsing experience. Tracking blockers may also cause missing visit data, so verify what happens when the page is actually opened.

Users reach the page but rarely take action: Focus on whether the ad promise matches the above-the-fold content. If the ad emphasizes fast delivery but the page says nothing about lead times, or users search for a specific model but the page only presents a company introduction, they may leave at the first step.

Users start filling out a form or checking out but do not complete it: Check required fields, phone number formats, error messages, shipping costs, and payment methods. In particular, check whether error messages on the French page are still in English and whether validation accepts commonly used local addresses and phone numbers.

If visit and action rates are similar in both languages, and the gap mainly occurs at the submission or payment stage, fixing functionality and transaction terms is more targeted than rewriting the entire set of ads.

Check whether localization communicates clearly, not just whether the translation is correct

Grammatically correct copy does not necessarily match local users’ search habits. First check that search terms, ad headlines, and product names on the page use consistent industry terminology. Then make sure the call-to-action clearly tells users what will happen next.

For example, an inquiry page should explain what users will receive after submitting the form and what information they need to provide. If the page only says “Contact us” without providing lead times, minimum order quantities, service coverage, or after-sales information, users may leave because they lack the information they need to make a decision. Online stores should display currency, delivery coverage, taxes and fees, and return policies in appropriate places.

Localization reviews should cover ads, above-the-fold content, forms, error messages, and post-submission pages. Translating only the main copy while leaving English buttons or unsuitable delivery information can still hurt conversions. Coverage in existing markets must also be described accurately; copy must not promise service capabilities that are not yet available.

Make changes based on evidence, then verify the results

Document each issue as “problem area—possible cause—validation action.” For example, if the completion rate for French mobile forms is low, fix field validation first. If French search terms do not reflect purchase intent, adjust keywords and negative keywords first. Address one primary hypothesis in each round and keep a record of changes. Avoid changing audiences, bids, copy, and landing pages all at once, as that makes it difficult to identify the cause.

Whenever possible, test pages using the same language and similar traffic conditions. Do not use French traffic from France and English traffic from the United States to directly evaluate copy performance. When reviewing business results, track qualified inquiries, subsequent sales, or order profit in addition to submission rates to avoid getting “more conversions, but less value.”

If different teams manage the ads and website, align conversion definitions and change logs. 易营宝 offers services such as multilingual website development and overseas ad campaigns. An integrated solution of this kind can be considered to coordinate the investigation; when evaluating providers, focus on whether they can jointly review ad sources, page experience, and actual business results.

As a next step, export data for both ad groups over the same period and complete the conversion journey once in French and once in English. After identifying the first step where the gap becomes clear, decide whether to adjust tracking, traffic, copy, or the landing page.