Lost in Translation? Language Localization Experiment and Its Impact on Conversions

NOTE: This is a raw transcript and contains grammatical errors. The curated transcript will be uploaded soon.

Paulius Zajančkauskas: Hello, everyone. I’m really

SCREENSHARE: happy to

Paulius Zajančkauskas: present here at Convex 2024. This is such an amazing

SCREENSHARE: opportunity to share

Paulius Zajančkauskas: and gain knowledge. So it’s great to be part this experience with you all. Now let me tell you a bit more about myself and how I’m joining on this journey. Uh, so my name is Paulus and I’m the leading the web growth team at Omniscient, where we are responsible for driving testing and experimentation strategies.

And our main goal is to activate visitors to sign up and convert them into paying customers by running various growth

SCREENSHARE: experiments.

Paulius Zajančkauskas: In

SCREENSHARE: today’s presentation, I will dive deeper

Paulius Zajančkauskas: into one of our recent experiments. It will

SCREENSHARE: include some

Paulius Zajančkauskas: theory to introduce practical aspects of the experimentation and how we are

SCREENSHARE: running at

Paulius Zajančkauskas: OmniSend. I will share more details about our localization experiment

SCREENSHARE: and its

Paulius Zajančkauskas: impact on our key metrics. Additionally, I will cover briefly our experimentation processes, and

SCREENSHARE: most important,

Paulius Zajančkauskas: If you stay till the end, I will present key considerations

SCREENSHARE: for

Paulius Zajančkauskas: other businesses that are trying to run similar

SCREENSHARE: experiment. And I will share

Paulius Zajančkauskas: exact framework that

SCREENSHARE: can use

Paulius Zajančkauskas: and adapt for your own needs. So make sure you don’t miss it.

SCREENSHARE: So firstly,

Paulius Zajančkauskas: how did we come up with the idea for localization testing?

SCREENSHARE: It is based

Paulius Zajančkauskas: on how we plan and set goals at Omnicent. We are not running a separate experimentation program with its own goals. But we look for the strategic opportunity and plan based for the, uh,

SCREENSHARE: based on

Paulius Zajančkauskas: the OKR framework.

SCREENSHARE: It

Paulius Zajančkauskas: is a goal

SCREENSHARE: setting framework used

Paulius Zajančkauskas: to define objectives and outcomes. So why do we use

SCREENSHARE: OKRs

Paulius Zajančkauskas: at all?

SCREENSHARE: It helps

Paulius Zajančkauskas: everyone to align and work

SCREENSHARE: towards the same goal.

Paulius Zajančkauskas: It helps to maintain focus as well. We have many ideas that we might be working on, but we need to focus on what truly matters. Uh, regular reviews allow you to track the progress and make changes when

SCREENSHARE: needed.

Paulius Zajančkauskas: And

SCREENSHARE: of course it

Paulius Zajančkauskas: involves employees in the goal setting processes. So it is increasing commitment and motivation to achieve the goals. Uh, I can share this exact Okia structure example that we’ve been working on that quarter.

SCREENSHARE: Uh, so first,

Paulius Zajančkauskas: the general

SCREENSHARE: direction

Paulius Zajančkauskas: and key results were set at the company

SCREENSHARE: level.

Paulius Zajančkauskas: Uh, each department then had an

SCREENSHARE: opportunity to

Paulius Zajančkauskas: evaluate. The situation and proposed potential key results. So in our case, it was a general direction to accelerate growth

SCREENSHARE: of new revenue and successfully expand in

Paulius Zajančkauskas: the dedicated market.

SCREENSHARE: Uh,

Paulius Zajančkauskas: therefore our team came up with the exact objective and key result to increase sign up

SCREENSHARE: rates

Paulius Zajančkauskas: and ultimately paying customers in that market. And after setting the

SCREENSHARE: exact

Paulius Zajančkauskas: key results, uh, that we’ll be focusing. So we start our idea generation processes,

SCREENSHARE: uh, to come up

Paulius Zajančkauskas: with a new ideas. We start

SCREENSHARE: standard brainstorming sessions,

Paulius Zajančkauskas: uh, in

SCREENSHARE: those sessions, we try to

Paulius Zajančkauskas: think creatively on

SCREENSHARE: how

Paulius Zajančkauskas: we can achieve our goals. In this stage, we don’t try to

SCREENSHARE: do exact measurement

Paulius Zajančkauskas: or validation of the ideas. We just try

SCREENSHARE: to generate as

Paulius Zajančkauskas: many ideas as possible with a clear

SCREENSHARE: explanation.

Paulius Zajančkauskas: Of course, we look back into the previous tests,

SCREENSHARE: tests that might be relevant.

Paulius Zajančkauskas: So we

SCREENSHARE: explore

Paulius Zajančkauskas: our research. And experimentation

SCREENSHARE: database. So it’s important

Paulius Zajančkauskas: to have a good documentation strategy before that. And after clear direction and key results are set, it gives you an indication on what data is worth checking.

SCREENSHARE: So in this case, we started to explore how different countries

Paulius Zajančkauskas: are performing in terms of signups and paying customers. So we tried to better understand what is the visitor behavior and how does it differ.

SCREENSHARE: The

Paulius Zajančkauskas: country as well.

SCREENSHARE: And

Paulius Zajančkauskas: one of the key idea that came up early was to make our offer easier to understand in the local market. We think that offering a translated experience will help visitors to better understand what OmniSend is all about, what value it offers, and what are the product capabilities.

SCREENSHARE: So

Paulius Zajančkauskas: in order to achieve that,

SCREENSHARE: we

Paulius Zajančkauskas: will be aiming to offer a translated

SCREENSHARE: experience for the whole customer journey.

Paulius Zajančkauskas: And this journey

SCREENSHARE: includes, uh,

Paulius Zajančkauskas: our main website, registration flow,

SCREENSHARE: and all information within our product.

Paulius Zajančkauskas: So, the visitor has the full experience and could quicker and better

SCREENSHARE: understand the value of Omniscent. Yeah. So based on

Paulius Zajančkauskas: idea generation efforts and

SCREENSHARE: our initial assumptions, we created

Paulius Zajančkauskas: a hypothesis

SCREENSHARE: that

Paulius Zajančkauskas: we are looking to validate further on. Uh, we believe that if we introduce a localized content in French for the France users, this will increase signups and ultimately paying customers because

SCREENSHARE: of increased trust

Paulius Zajančkauskas: and comfort. Uh, we also

SCREENSHARE: believe that people generally prefer browsing websites and using

Paulius Zajančkauskas: products in their local language, which

SCREENSHARE: enhance the

Paulius Zajančkauskas: overall appeal of our offering as well. Yeah,

SCREENSHARE: so our next

Paulius Zajančkauskas: step was to select the hypothesis that we will be looking to experiment with further on. Uh, before

SCREENSHARE: diving

Paulius Zajančkauskas: deeper into

SCREENSHARE: prioritization,

Paulius Zajančkauskas: we first use an assumption

SCREENSHARE: map

Paulius Zajančkauskas: to position our

SCREENSHARE: ideas and

Paulius Zajančkauskas: hypothesis. We start by

SCREENSHARE: assuming which of our

Paulius Zajančkauskas: ideas are more important for reaching the planned goals.

SCREENSHARE: We don’t aim for

Paulius Zajančkauskas: high accuracy

SCREENSHARE: here,

Paulius Zajančkauskas: but rather

SCREENSHARE: evaluate

Paulius Zajančkauskas: if it fits our general direction

SCREENSHARE: and if it will help to achieve our goals

Paulius Zajančkauskas: as well. Next, uh, we assess how much evidence we have,

SCREENSHARE: And our focus

Paulius Zajančkauskas: area is on the hypotheses that are placed on the top right

SCREENSHARE: corner.

Paulius Zajančkauskas: So it is indicating that we already have some light

SCREENSHARE: evidence supporting the idea,

Paulius Zajančkauskas: but

SCREENSHARE: we still

Paulius Zajančkauskas: need more research and validation. If we identify ideas that are highly important

SCREENSHARE: and already are supported by strong evidence, so of

Paulius Zajančkauskas: course our

SCREENSHARE: validation process is much quicker and we can implement such

Paulius Zajančkauskas: ideas faster.

SCREENSHARE: However,

Paulius Zajančkauskas: this was not the case with our localization experiment and we needed to experiment further.

SCREENSHARE: Uh,

Paulius Zajančkauskas: We do the final

SCREENSHARE: prioritization based on

Paulius Zajančkauskas: the standardized framework.

SCREENSHARE: I

Paulius Zajančkauskas: believe most of our listeners already know this framework, so just a quick reminder that we firstly evaluate hypothesis impact

SCREENSHARE: on our key metric.

Paulius Zajančkauskas: Uh,

SCREENSHARE: then we

Paulius Zajančkauskas: Estimate how much effort and resources I needed to implement it.

SCREENSHARE: and finally,

Paulius Zajančkauskas: we have a light. How confident we are about the impact and ease

SCREENSHARE: of the

Paulius Zajančkauskas: implementation.

SCREENSHARE: and we

Paulius Zajančkauskas: dive a bit deeper into the confidence level. Uh, and we try to, uh, avoid making updates just based on their opinions.

SCREENSHARE: So minor tweaks and changes could be done by

Paulius Zajančkauskas: just

SCREENSHARE: by

Paulius Zajančkauskas: assessing them. And if the idea is scoring low, we are looking for the ways on how we can increase their confidence and gather more

SCREENSHARE: evidence to support the

Paulius Zajančkauskas: idea. It’s true that sometimes depending on the specific idea,

SCREENSHARE: we

Paulius Zajančkauskas: tolerate higher risk

SCREENSHARE: and

Paulius Zajančkauskas: we might implement Ideas that have low or medium confidence as well. Yeah, it is. Time to start our

SCREENSHARE: validation process

Paulius Zajančkauskas: and see if our direction is correct. So in general,

SCREENSHARE: when

Paulius Zajančkauskas: starting to explore the new idea,

SCREENSHARE: there is a lot of uncertainty

Paulius Zajančkauskas: in the beginning. Lots of

SCREENSHARE: are not clear. Uh, there is a lot of

Paulius Zajančkauskas: confusion, uh, how

SCREENSHARE: it might impact

Paulius Zajančkauskas: our business.

SCREENSHARE: Can it move the needle at

Paulius Zajančkauskas: all in our key

SCREENSHARE: metric and what resources

Paulius Zajančkauskas: might be needed? So transforming

SCREENSHARE: an idea

Paulius Zajančkauskas: into a successful

SCREENSHARE: experiment,

Paulius Zajančkauskas: it requires additional effort

SCREENSHARE: as well.

Paulius Zajančkauskas: Usually from initial ideas, we move to

SCREENSHARE: search and testing stage. We firstly focus on the discovery part that includes exploration, whether the

Paulius Zajančkauskas: general direction is

SCREENSHARE: right.

Paulius Zajančkauskas: It includes testing basic assumptions and gathering more insights if it’s worth to continue the experiment. Uh, yeah, validation stage focuses on testing the direction

SCREENSHARE: you

Paulius Zajančkauskas: have taken. It

SCREENSHARE: involves getting stronger evidences that

Paulius Zajančkauskas: the idea is more likely

SCREENSHARE: to succeed. And after the validation, uh, we moved to the execution phase where experiment is implemented,

Paulius Zajančkauskas: implemented and will likely make a positive impact on overall business. Uh, how does the validation process looked in, in our way?

SCREENSHARE: So we started our discovery stage by analyzing

Paulius Zajančkauskas: our website

SCREENSHARE: visitors browsers

Paulius Zajančkauskas: language.

SCREENSHARE: This gave us indication

Paulius Zajančkauskas: how much of the

SCREENSHARE: visitors are using their local language. Uh, secondly, we implemented additional tracking to measure

Paulius Zajančkauskas: how much of the website visitors are

SCREENSHARE: translating our website into their

Paulius Zajančkauskas: local

SCREENSHARE: language.

Paulius Zajančkauskas: Our validation stage

SCREENSHARE: uh,

Paulius Zajančkauskas: had three steps, and it started with a paid search

SCREENSHARE: campaign

Paulius Zajančkauskas: where we A B tested translated ads and its

SCREENSHARE: impact. As the

Paulius Zajančkauskas: next step for the

SCREENSHARE: web visitors,

Paulius Zajančkauskas: we have introduced

SCREENSHARE: an option to browse the website in their preferred

Paulius Zajančkauskas: language. Which was English

SCREENSHARE: or

Paulius Zajančkauskas: French

SCREENSHARE: in our case.

Paulius Zajančkauskas: And actually, this was the

SCREENSHARE: key

Paulius Zajančkauskas: step in our validation process. It helped to get real

SCREENSHARE: usage data and

Paulius Zajančkauskas: get additional qualitative

SCREENSHARE: insights

Paulius Zajančkauskas: as well. And as the final step, we have

SCREENSHARE: analyzed

Paulius Zajančkauskas: the impact to our

SCREENSHARE: signups and how we are getting

Paulius Zajančkauskas: and how much new customers we are

SCREENSHARE: getting. Uh, before

Paulius Zajančkauskas: executing the experiment, it required to align

SCREENSHARE: between

Paulius Zajančkauskas: several teams. Uh,

SCREENSHARE: we needed

Paulius Zajančkauskas: to include

SCREENSHARE: digital ads team

Paulius Zajančkauskas: that was responsible for launching dedicated paid search

SCREENSHARE: campaign, content and design

Paulius Zajančkauskas: teams help to build translated website. Uh, the

SCREENSHARE: product team was responsible for the

Paulius Zajančkauskas: translation of the

SCREENSHARE: sign

Paulius Zajančkauskas: up flow and all content within our product.

SCREENSHARE: And what is more,

Paulius Zajančkauskas: the data ops team was included from the beginning of the experiment. So we make sure we track everything

SCREENSHARE: correctly

Paulius Zajančkauskas: for the final measurements and conclusions. Our team acted as a leading team in experiment

SCREENSHARE: execution

Paulius Zajančkauskas: And aligning everyone for the success. So how

SCREENSHARE: do we align everyone?

Paulius Zajančkauskas: At least in our

SCREENSHARE: case, we

Paulius Zajančkauskas: have succeeded with simple three steps.

SCREENSHARE: So we gathered

Paulius Zajančkauskas: for the initial kickoff meeting where I presented the initial experiment idea, the logic behind it, and why it is important not

SCREENSHARE: only for our team,

Paulius Zajančkauskas: but reaching wider organization goals.

SCREENSHARE: also shared

Paulius Zajančkauskas: the detailed

SCREENSHARE: plan. What is expected

Paulius Zajančkauskas: from each of the after sharing the initial idea. It is important to get feedback on the experiment execution.

SCREENSHARE: So other teams engagement could not

Paulius Zajančkauskas: only shape experiment for success, but also increase the sense of involvement.

SCREENSHARE: And yeah,

Paulius Zajančkauskas: during the

SCREENSHARE: preparation

Paulius Zajančkauskas: stage and while experiment is running, so it is important to keep

SCREENSHARE: everyone in

Paulius Zajančkauskas: the loop and share what is

SCREENSHARE: happening with the experiment,

Paulius Zajančkauskas: share its

SCREENSHARE: important milestones so

Paulius Zajančkauskas: all teams are aware of the situation and could make necessary changes

SCREENSHARE: if

Paulius Zajančkauskas: needed. So now

SCREENSHARE: let’s see what

Paulius Zajančkauskas: we have identified in

SCREENSHARE: our

Paulius Zajančkauskas: discovery stage. So we started to analyze how users

SCREENSHARE: with different browser language settings.

Paulius Zajančkauskas: Convert to registrations. This data which is

SCREENSHARE: tracked by default in GA4 and comes primarily from Chrome browsers, revealed some

Paulius Zajančkauskas: interesting insights. We noticed that visitors with a French language convert twice less compared

SCREENSHARE: to the ones in English.

Paulius Zajančkauskas: This gave us indication that there is an

SCREENSHARE: opportunity to improve our offering in their local language.

Paulius Zajančkauskas: Additionally, the French language accounted for three times more signups compared to other closest languages, and French was the Most

SCREENSHARE: popular browser language after English.

Paulius Zajančkauskas: Uh, during the previous

SCREENSHARE: research, we noticed that

Paulius Zajančkauskas: is a bunch of visitors that are coming to our website, and they are using a Google Translate functionality in order to translate the website. Therefore, to get

SCREENSHARE: additional

Paulius Zajančkauskas: data,

SCREENSHARE: we have implemented additional translation

Paulius Zajančkauskas: tracking. As we can see from the table,

SCREENSHARE: most of the

Paulius Zajančkauskas: translations were initiated from France.

SCREENSHARE: It accounts

Paulius Zajančkauskas: for nearly 10 percent of all

SCREENSHARE: translations.

Paulius Zajančkauskas: This means that local content is more important for French users compared to users from any other country. As the next step, we wanted

SCREENSHARE: to gather more evidences.

Paulius Zajančkauskas: In order to bring

SCREENSHARE: more high intent visitors,

Paulius Zajančkauskas: we have launched a dedicated paid

SCREENSHARE: search

Paulius Zajančkauskas: campaign. We have chosen to use a paid

SCREENSHARE: search campaign

Paulius Zajančkauskas: based on the previous Campaign performance and we know the ability that it can bring us additional high intent signups to our product. We

SCREENSHARE: have also tested different languages in

Paulius Zajančkauskas: the ads to measure its impact and

SCREENSHARE: engagement

Paulius Zajančkauskas: and original campaign showed search

SCREENSHARE: ads in English

Paulius Zajančkauskas: and

SCREENSHARE: the trial campaign showed ads in

Paulius Zajančkauskas: French.

SCREENSHARE: And this was launched as

Paulius Zajančkauskas: an A B test in the Google ads platform as the second step in our validation journey. We introduced a localized

SCREENSHARE: experience selection

Paulius Zajančkauskas: for the visitors that are

SCREENSHARE: coming from France.

Paulius Zajančkauskas: So we translated 15 key pages that were crucial in

SCREENSHARE: the

Paulius Zajančkauskas: visitor conversion path. Uh,

SCREENSHARE: the experience included the translated sign up flow as

Paulius Zajančkauskas: well, and we implemented

SCREENSHARE: additional built in service in the experience

Paulius Zajančkauskas: selection. Of

SCREENSHARE: course, all of the localization experience

Paulius Zajančkauskas: had full tracking, so we have implemented event tracking on key actions,

SCREENSHARE: but also used URL parameters

Paulius Zajančkauskas: to fully track and identify the users.

SCREENSHARE: So,

Paulius Zajančkauskas: all of this helped us get early insights from our live users and identify their preferred behavior as well.

SCREENSHARE: So, it’s time to see

Paulius Zajančkauskas: the results that we have achieved. So, let’s take a closer look. Just a reminder that we were A

SCREENSHARE: B

Paulius Zajančkauskas: testing English and French ads.

SCREENSHARE: So, this is kind

Paulius Zajančkauskas: of obvious, but 94

SCREENSHARE: of the

Paulius Zajančkauskas: visitors coming through French ads

SCREENSHARE: chose to continue their experience in

Paulius Zajančkauskas: French.

SCREENSHARE: What

Paulius Zajančkauskas: is a bit more

SCREENSHARE: interesting,

Paulius Zajančkauskas: that nearly

SCREENSHARE: 70

Paulius Zajančkauskas: percent of the visitors that engaged

SCREENSHARE: with English

Paulius Zajančkauskas: ads

SCREENSHARE: chose to continue their journey in French.

Paulius Zajančkauskas: So

SCREENSHARE: it seems that at least

Paulius Zajančkauskas: in our case target audience is using general terminology in English however,

SCREENSHARE: when they

Paulius Zajančkauskas: have

SCREENSHARE: the

Paulius Zajančkauskas: possibility Majority of them try to use the local language And, in addition, the trial campaign with the translated ads showed 73 percent

SCREENSHARE: higher sign up rate.

Paulius Zajančkauskas: And this resulted in

SCREENSHARE: 51 percent lower cost per sign

Paulius Zajančkauskas: up. So,

SCREENSHARE: that

Paulius Zajančkauskas: was a great result. When looking at all users, no matter their acquisition channel,

SCREENSHARE: the majority of visitors have chosen to browse our website in

Paulius Zajančkauskas: the local language, French.

SCREENSHARE: and

Paulius Zajančkauskas: a bit more than 20 percent of the

SCREENSHARE: visitors

Paulius Zajančkauskas: We’re still interested in using English and we

SCREENSHARE: can

Paulius Zajančkauskas: the final results of the experiment. So with the local experience, we were able to achieve 35 percent uplift in our

SCREENSHARE: sign up rate. English version sign up rate

Paulius Zajančkauskas: was nine, 9. 5, whereas the French experience resulted in nearly 13 percent sign up

SCREENSHARE: rate.

Paulius Zajančkauskas: I know a

SCREENSHARE: big

Paulius Zajančkauskas: part of our audience might be already

SCREENSHARE: questioning what is the confidence of such results.

Paulius Zajančkauskas: So in our case,

SCREENSHARE: differences

Paulius Zajančkauskas: in the sign up

SCREENSHARE: rate were statistically significant.

Paulius Zajančkauskas: and we were

SCREENSHARE: able

Paulius Zajančkauskas: to state that

SCREENSHARE: with a high

Paulius Zajančkauskas: confidence level of

SCREENSHARE: 95%.

Paulius Zajančkauskas: And yeah, even though the activation rate

SCREENSHARE: was almost

Paulius Zajančkauskas: the

SCREENSHARE: same with only minor

Paulius Zajančkauskas: differences,

SCREENSHARE: the most

Paulius Zajančkauskas: important outcome, uh, that we were able

SCREENSHARE: to

Paulius Zajančkauskas: bring more paying customers. uh, however, due to the

SCREENSHARE: long customer buying buying cycle, we needed more paying users

Paulius Zajančkauskas: to fully validate that

SCREENSHARE: as

Paulius Zajančkauskas: well.

SCREENSHARE: Yeah. And

Paulius Zajančkauskas: so after

SCREENSHARE: all

Paulius Zajančkauskas: experiment analysis,

SCREENSHARE: uh, we have made the following

Paulius Zajančkauskas: conclusions. So the

SCREENSHARE: the majority of the users from France

Paulius Zajančkauskas: prefer to use the

SCREENSHARE: local language when discovering

Paulius Zajančkauskas: new products and browsing websites.

SCREENSHARE: Uh, our acquisition

Paulius Zajančkauskas: campaigns are more effective. Resulting in significant

SCREENSHARE: uplift in sign up rates

Paulius Zajančkauskas: and reducing cost per

SCREENSHARE: acquisition.

Paulius Zajančkauskas: and we

SCREENSHARE: can confidently say that the local

Paulius Zajančkauskas: experience helped to activate visitors and of course,

SCREENSHARE: making them more likely

Paulius Zajančkauskas: to sign up to our product and all of these optimized Thank you. Touch points resulted in attracting more paying customers. So

SCREENSHARE: as mentioned earlier, it’s time to

Paulius Zajančkauskas: share exact validation framework we used to test our localization efforts. So I’m

SCREENSHARE: excited to finally show it

Paulius Zajančkauskas: to you. This framework is easy to understand and can be

SCREENSHARE: adapted

Paulius Zajančkauskas: to

SCREENSHARE: fit your own needs.

Paulius Zajančkauskas: Of

SCREENSHARE: every

Paulius Zajančkauskas: situation

SCREENSHARE: different,

Paulius Zajančkauskas: so feel free

SCREENSHARE: to

Paulius Zajančkauskas: tweak it based on your

SCREENSHARE: niche, company stage,

Paulius Zajančkauskas: or your growth goals.

SCREENSHARE: Thanks.

Paulius Zajančkauskas: So you need to start your localization efforts by identifying key

SCREENSHARE: acquisitional channels that are working well for you.

Paulius Zajančkauskas: It’s important

SCREENSHARE: that you can confidently

Paulius Zajančkauskas: state that these channels are bringing high quality signups and paying customers to you.

SCREENSHARE: This way you

Paulius Zajančkauskas: will reduce external factors that influence your campaign performance and you will be more likely to attribute the success or failure

SCREENSHARE: of your localization

Paulius Zajančkauskas: efforts. Use that channel to A, B test translated ads. Uh, then get Real visitor engagement data by offering to use that translated version of your website. Focus on building

SCREENSHARE: MVP version.

Paulius Zajančkauskas: analyze and identify

SCREENSHARE: key pages in the user conversion

Paulius Zajančkauskas: path. Use an opportunity to

SCREENSHARE: build,

Paulius Zajančkauskas: to use built

SCREENSHARE: in surveys.

Paulius Zajančkauskas: So at

SCREENSHARE: that same

Paulius Zajančkauskas: moment, you will be able to know why they are making a specific action

SCREENSHARE: and

Paulius Zajančkauskas: are the visitor motivation behind that.

SCREENSHARE: And make sure

Paulius Zajančkauskas: that visitors are able to switch

SCREENSHARE: between different

Paulius Zajančkauskas: languages and between different

SCREENSHARE: experiences as well. And you need to

Paulius Zajančkauskas: look into the whole customer journey. If the

SCREENSHARE: customer journey is long

Paulius Zajančkauskas: and there are more touch points, make sure that the new localized experience is

SCREENSHARE: unified across them until

Paulius Zajančkauskas: the user is converted. When running a localization experiment, businesses should consider the

SCREENSHARE: following.

Paulius Zajančkauskas: Think. Uh, test localization across acquisition channels, not

SCREENSHARE: just

Paulius Zajančkauskas: on your own website. Uh, this

SCREENSHARE: will provide insights whether localization

Paulius Zajančkauskas: improves your

SCREENSHARE: acquisition.

Paulius Zajančkauskas: Uh,

SCREENSHARE: optimize first

Paulius Zajančkauskas: customer touch points with

SCREENSHARE: a local experience.

Paulius Zajančkauskas: So this can significantly improve your acquisition efforts. You need

SCREENSHARE: get

Paulius Zajančkauskas: creative with your validation process.

SCREENSHARE: explore

Paulius Zajančkauskas: early signals before moving to the, to the validation

SCREENSHARE: stage. And yes,

Paulius Zajančkauskas: language is only one aspect of localization. Other factors such as value proposition, messaging should be considered as

SCREENSHARE: well.

Paulius Zajančkauskas: And of

SCREENSHARE: course, you need to

Paulius Zajančkauskas: prioritize

SCREENSHARE: your

Paulius Zajančkauskas: cross team collaboration so to ensure alignment and open communication across involved teams. So this shapes the experiment to this

SCREENSHARE: success. Yeah, and this

Paulius Zajančkauskas: entire experiment was driven

SCREENSHARE: by the leading

Paulius Zajančkauskas: web growth team. So I’d like to take

SCREENSHARE: this opportunity to

Paulius Zajančkauskas: give a shout out to

SCREENSHARE: other core team

Paulius Zajančkauskas: members who helped not only,

SCREENSHARE: uh, implement this successful experiment,

Paulius Zajančkauskas: but also built the

SCREENSHARE: Experimentation

Paulius Zajančkauskas: framework

SCREENSHARE: that we are already for already using for other tests

Paulius Zajančkauskas: in our company. So thanks everyone once

SCREENSHARE: again. Uh,

Paulius Zajančkauskas: If you have any questions or suggestions, feel free to reach

SCREENSHARE: out

Paulius Zajančkauskas: to me. I’m happy to connect and chat

SCREENSHARE: about any growth and experimentation topics.

Paulius Zajančkauskas: So your

SCREENSHARE: insights and feedback always welcome as well.

Paulius Zajančkauskas: So hope you found

SCREENSHARE: this presentation interesting.

Paulius Zajančkauskas: Uh, enjoy the rest

SCREENSHARE: the conference. Uh, thank

Paulius Zajančkauskas: you. And bye bye.