Feeding Your Dashboard with Quality Information

We collect the data you need to grow your brand so you don't have to. Our data collection process is not a secret - let us tell you how we do it.

Where We Get Our Data

Latana sits under the Dalia Research umbrella. Dalia is a powerful survey engine for real-time market and opinion research. What is amazing about Dalia is that its platform can be integrated into anything from freemium apps to web advertisements and custom target approaches. As a result of these integrations,

Dalia has instant access to 40,000 apps and websites. These apps range from gaming to utilities, toentertainment, health and beyond, meaning we have excellent reach across all demographics. Still with us, good? Dalia leverages these apps and websites to be able to distribute 5 million micro-surveys per month to people living in over 100 countries. Due to the massive variety of apps and websites we appear on, we are easily able to reach a multitude of demographics.

It lends its technology to us so that we can deliver the essential insights your brand needs to thrive. You are in good hands. Stanford University, Millward Brown, and The Guardian newspaper all use Dalia’s data too.

Online Sampling

Dalia basically does online sampling on mobile devices. We will get to the aspect of the mobile devices later on. For now, let’s focus on online sampling and, more specifically, why Dalia uses global sampling.

The overarching reason is that it allows connection to the 50%+ of the global population that is connected to the internet. This makes for a very representative and efficient channel for opinion research, especially in terms of cost and speed. However, the advantages don’t stop there. Online sampling also allows for:

Dalia basically does online sampling on mobile devices. We will get to the aspect of the mobile devices later on. For now, let’s focus on online sampling and, more specifically, why Dalia uses global sampling.

The overarching reason is that it allows connection to the 50%+ of the global population that is connected to the internet. This makes for a very representative and efficient channel for opinion research, especially in terms of cost and speed. However, the advantages don’t stop there. Online sampling also allows for:

Latana tick icon

Consistent sampling

Latana tick icon

Comparison of data across countries

Latana tick icon

A reduction in bias thanks to anonymity

Latana tick icon

Trustworthy data

Latana tick icon

Higher participation

Online Sampling on Mobile Devices

More and more people are accessing the internet via a smartphone than a desktop computer. This makes it easier to seamlessly mobile surveys into everyday life. This is especially true in emerging markets, the markets that have traditionally been the hardest to reach.

How the Survey System Works

Dalia utilizes its partnership with publishers i.e. companies that own apps or websites and/or operate large networks of apps or websites. Our network includes any app or website you can think of from gaming to utilities and beyond. Surveys are placed on the apps or websites of the publishers, where their users are incentivized to complete the survey. These incentives can be monetary or non-monetary (e.g. receive 20 gold coins on a gaming app in return for a survey completion). In turn, the publisher is paid every time a user completes a survey.

Where We Get Users From

Our users come from numerous apps and websites worldwide. Here are some examples.

Yuno

One example is our very own online survey panel, Yuno. Yuno is advertised via Google and Facebook, or can be found online at https://yunosurveys.com/signup. Users sign up to Yuno using their email address and phone number and from there on, complete surveys.

LINE

Line is a messaging app that has over 217 million monthly active users and is one of the most highly used messaging apps in the world.

Line offers “LINE Points”, a service through which users can earn points and virtual currency by watching content, taking surveys, and adding official accounts as friends. With this feature, users can access Latana’s surveys and gain points for completing the survey.

Focus on User Experience

Concentration is on good user experience so high-quality survey data can be received. Several elements make up a good user experience:

Latana tick icon

A fun design

Latana tick icon

An engaging survey

Latana tick icon

Short surveys that have under 200 characters per question

Latana tick icon

Avoiding long lists of answer options

Latana tick icon

Option not to answer questions

Latana tick icon

Professional translators to verify localization


When there is great user experience, respondents tend to provide accurate, thoughtful, and honest input.

Data Quality

Gathering survey respondents isn’t enough. We take a step further to ensure you receive only quality data. We do this by dynamically profiling each potential respondent across key demographic, psychographic, and behavioural attributes.

We are able to assign a persistent identifier to every user. They do this by:

Latana tick icon

Using digital fingerprinting and device identifier technologies

Latana tick icon

Blacklisting low-quality sources

Latana tick icon

Tracking over 100 active and passive profile parameters

Latana tick icon

Automating in-survey behaviour checks

Latana tick icon

Using adaptive screening based on user behavior


Then, using machine-learning algorithms that analyse active and passive data, we generate a quality and trust score for each potential respondent. 

An audience attribution engine matches qualified respondents to a survey targeting their profile. Sounds pretty tech-heavy, and it is, but it guarantees that only engaged and accomplished people complete each survey.

Multiple Completes

Parameters are also put in place to ensure that users can’t participate in the same survey more than once. This is done using an identifier based on cookies in a user's browser cache. 

Multiple Completes

We work to avoid bad responses by using a real-time and adaptive algorithm to track, analyse and profile respondents based on three behavioral components: consistency, reliability and attention. These take into account:

Latana tick icon

Speed of response

Latana tick icon

Inconsistent answers

Latana tick icon

Click behaviour

Latana tick icon

Bad open-ended answer

Where We Control For Real World Changes

Where We Control For Real World Changes

Where We Control For Real World Changes

We don’t do random. We do precise and accurate. Therefore, any changes you might see are based on real-world changes because we take the time to control our sample for a large number of variables. We’ve invited a few friends to show how controlling works. In traditional quota sampling, usually just age and gender are controlled to maintain sample composition. 
Let’s take this as an example:

A survey of 1,000 people with the following distribution, in terms of demographics.

500 Male / 500 Female

Age and gender are controlled so the researcher will make sure that the ratio of both is maintained in the next survey. This is to ensure the results are comparable. Bear in mind that only these variables will be controlled. Let’s keep this information in mind and move onto the next survey. Here, the sample composition dramatically shifts towards urban location:

500 Male / 500 Female (stable, because controlled)
600 young / 400 old (stable, because controlled)
800 urban / 200 rural (skewed, because not controlled for)

The demographic groups in the first survey are fundamentally different from each other. Because of this, it is now likely that you will see changes in the results because of the higher proportion of the urban population in the sample. 

This is important because, particularly when it comes to brand perception, such demographic variables can have a great impact on brand performance.

So, what does it mean for you when you receive insights from surveys where the minimum variables have been controlled? Misinformation. 
The insights are not based on real-world changes but rather changes in sample composition. Basing decisions on inaccurate insights can be dangerous for your brand.

Where We Get Our Data

Latana sits under the Dalia Research umbrella. Dalia is a powerful survey engine for real-time market and opinion research. What is amazing about Dalia is that its platform can be integrated into anything from freemium apps to web advertisements and custom target approaches.

+ Read more

Dalia has instant access to 40,000 apps and websites. These apps range from gaming to utilities, toentertainment, health and beyond, meaning we have excellent reach across all demographics. Still with us, good? Dalia leverages these apps and websites to be able to distribute 5 million micro-surveys per month to people living in over 100 countries. Due to the massive variety of apps and websites we appear on, we are easily able to reach a multitude of demographics.

It lends its technology to us so that we can deliver the essential insights your brand needs to thrive. You are in good hands. Stanford University, Millward Brown, and The Guardian newspaper all use Dalia’s data too.

Online Sampling

Dalia basically does online sampling on mobile devices. We will get to the aspect of the mobile devices later on. For now, let’s focus on online sampling and, more specifically, why Dalia uses global sampling.

+ Read more

Dalia basically does online sampling on mobile devices. We will get to the aspect of the mobile devices later on. For now, let’s focus on online sampling and, more specifically, why Dalia uses global sampling.

The overarching reason is that it allows connection to the 50%+ of the global population that is connected to the internet. This makes for a very representative and efficient channel for opinion research, especially in terms of cost and speed. However, the advantages don’t stop there. Online sampling also allows for:

Latana tick icon

Consistent sampling

Latana tick icon

Comparison of data across countries

Latana tick icon

A reduction in bias thanks to anonymity

Latana tick icon

Trustworthy data

Latana tick icon

Higher participation

Online Sampling on Mobile Devices

More and more people are accessing the internet via a smartphone than a desktop computer. This makes it easier to seamlessly mobile surveys into everyday life. This is especially true in emerging markets, the markets that have traditionally been the hardest to reach.

How the Survey System Works

Dalia utilizes its partnership with publishers i.e. companies that own apps or websites and/or operate large networks of apps or websites. Our network includes any app or website you can think of from gaming to utilities and beyond. Surveys are placed on the apps or websites of the publishers, where their users are incentivized to complete the survey. These incentives can be monetary or non-monetary (e.g. receive 20 gold coins on a gaming app in return for a survey completion). In turn, the publisher is paid every time a user completes a survey.

Where We Get Users From

Our users come from numerous apps and websites worldwide. Here are some examples.

Yuno

One example is our very own online survey panel, Yuno. Yuno is advertised via Google and Facebook, or can be found online at https://yunosurveys.com/signup. Users sign up to Yuno using their email address and phone number and from there on, complete surveys.

LINE

Line is a messaging app that has over 217 million monthly active users and is one of the most highly used messaging apps in the world.

Line offers “LINE Points”, a service through which users can earn points and virtual currency by watching content, taking surveys, and adding official accounts as friends. With this feature, users can access Latana’s surveys and gain points for completing the survey.

Focus on User Experience

Concentration is on good user experience so high-quality survey data can be received. Several elements make up a good user experience:

+ Read more
Latana tick icon

A fun design

Latana tick icon

An engaging survey

Latana tick icon

Short surveys that have under 200 characters per question

Latana tick icon

Avoiding long lists of answer options

Latana tick icon

Option not to answer questions

Latana tick icon

Professional translators to verify localization

When there is great user experience, respondents tend to provide accurate, thoughtful, and honest input.

Data Quality

Gathering survey respondents isn’t enough. We take a step further to ensure you receive only quality data. We do this by dynamically profiling each potential respondent across key demographic, psychographic, and behavioural attributes.

+ Read more

Gathering survey respondents isn’t enough. We take a step further to ensure you receive only quality data. We do this by dynamically profiling each potential respondent across key demographic, psychographic, and behavioural attributes.

We are able to assign a persistent identifier to every user. They do this by:

Latana tick icon

Using digital fingerprinting and device identifier technologies

Latana tick icon

Blacklisting low-quality sources

Latana tick icon

Tracking over 100 active and passive profile parameters

Latana tick icon

Automating in-survey behaviour checks

Latana tick icon

Using adaptive screening based on user behavior

Then, using machine-learning algorithms that analyse active and passive data, we generate a quality and trust score for each potential respondent. 

An audience attribution engine matches qualified respondents to a survey targeting their profile. Sounds pretty tech-heavy, and it is, but it guarantees that only engaged and accomplished people complete each survey.

Multiple Completes

Parameters are also put in place to ensure that users can’t participate in the same survey more than once. This is done using an identifier based on cookies in a user's browser cache. 

We work to avoid bad responses by using a real-time and adaptive algorithm to track, analyse and profile respondents based on three behavioral components: consistency, reliability and attention. These take into account:

Latana tick icon

Speed of response

Latana tick icon

Inconsistent answers

Latana tick icon

Click behaviour

Latana tick icon

Bad open-ended answer

Where We Control For Real World Changes

We don’t do random. We do precise and accurate. Therefore, any changes you might see are based on real-world changes because we take the time to control our sample for a large number of variables.

+ Read more

We don’t do random. We do precise and accurate. Therefore, any changes you might see are based on real-world changes because we take the time to control our sample for a large number of variables. We’ve invited a few friends to show how controlling works. In traditional quota sampling, usually just age and gender are controlled to maintain sample composition. 
Let’s take this as an example:

A survey of 1,000 people with the following distribution, in terms of demographics.

500 Male / 500 Female

Age and gender are controlled so the researcher will make sure that the ratio of both is maintained in the next survey. This is to ensure the results are comparable. Bear in mind that only these variables will be controlled. Let’s keep this information in mind and move onto the next survey. Here, the sample composition dramatically shifts towards urban location:

500 Male / 500 Female (stable, because controlled)
600 young / 400 old (stable, because controlled)
800 urban / 200 rural (skewed, because not controlled for)

The demographic groups in the first survey are fundamentally different from each other. Because of this, it is now likely that you will see changes in the results because of the higher proportion of the urban population in the sample. 

This is important because, particularly when it comes to brand perception, such demographic variables can have a great impact on brand performance.

So, what does it mean for you when you receive insights from surveys where the minimum variables have been controlled? Misinformation. 
The insights are not based on real-world changes but rather changes in sample composition. Basing decisions on inaccurate insights can be dangerous for your brand.

Cartoon depiction of insights