The Application of Coding in Market Research

When working in high-volume market research projects, finding meaningful insights and picking up on trends among hundreds of responses can sometimes be like finding a needle in a haystack. And when the research spans territories and comes in a range of different languages, the task can be even more challenging. 
 
At a time in which organisations and businesses alike are leveraging digital opportunities to expand their reach and operate across markets agilely, being able to gain insight into consumers’ choices and opinions quickly and efficiently has become a key priority.  
 
In this context, where businesses want to be always one step ahead, coding continues to be a fundamental instrument to help market researchers do exactly that: make vast troves of data more manageable and easier to analyse while being cost and time efficient. But how exactly does coding work? And what role does translation play in the coding process? 
 
Unearthing trends efficiently 
Open-ended responses in market research are one of the most powerful ways brands and companies can obtain insight from consumers. With this method, they can provide feedback in their own words, focusing on the aspects that most matter to them rather than being constrained by previously set options. However, this can lead to an increased difficulty when it comes to drawing conclusions. 
 
Coding groups similar responses into different categories and assigns them a specific code. For example, in an online survey whose objective it is to gather participants’ reactions to a new cleaning product, coders may classify comments such as “removes all stains effectively”, “great for cleaning all surfaces” and “powerful cleaning product” under the code for “effective”. If the research is multi-market, coding in the source language will be essential to capture the nuances of each response and will allow for a more accurate assigning of codes. This is why THG Fluently works with native coders and linguists who are carefully selected for each project based on their industry expertise. 
 
Once the coding has been completed, researchers can understand how the product fared among consumers by using the codes for quantitative data analysis, rather than reading through hundreds (or thousands) of individual responses. Working with the grouped responses cuts down considerably on the time and resources needed to unearth emerging trends and relevant data points, while giving researchers the possibility to also look at each response in detail to gain a deeper understanding of what drives common themes and prominent comments.  
 
For coding to yield positive results, the process needs to be tailored to the project’s requirements and main objectives. First, a code frame will need to be set. This will include the list of expected (or most common) categories, to which a unique code will be assigned. Other parameters also need to be clarified before starting, such as in which cases to create new categories, subcategories, nets and codes (or how many similar responses are needed to create a new code), and the criteria to apply the code for an invalid or irrelevant response.  
 
Automation: augmented efficiency throughout the process 
Coding is a useful tool within the overall market research data analysis process, as it provides an idea of consumers’ opinions in a more straightforward, quicker way. However, it can still be a time-intensive task for coders, who need to go through the individual responses to categorise them as needed. 
 
Nowadays, coding platforms enhanced with artificial intelligence (AI) capabilities can support and augment the coder’s work by, for instance, highlighting relevant words or expressions that will help decode the sentiment of a response, thus making it easier to categorise. Some systems can also suggest codes, therefore reducing the amount of time needed to process a batch of responses. For multilingual projects, some platforms can also offer a machine translation function, so that the coding and translation of the responses take place in the same environment. The translation can then be reviewed and edited accordingly by human linguists to improve quality. 
 
The target audience, whether they be individuals or fellow companies, are at the heart of any business, and market research is an invaluable aid to get insight into their thoughts and opinions. At a time in which organisations want to move faster than ever to win over foreign markets, coding is instrumental to quickly learn more about new audiences in their own words — what they want, what they enjoy, and what they dislike. Working with their native language, rather than translating the data first and then coding it, will not only ensure that the analysis is accurate, but also will make the process highly efficient and less resource intensive. 
 
Coding and translation can go hand-in-hand to ensure global market research is more efficient and accessible than ever, but it's not the only technology that can help create efficiencies. In our full report on The Technology Easing International Market Research, we take a deep dive into the benefits and use cases for not only coding, but also machine translation and voice-to-text technology. Access the full article here. 
 
At THG Fluently, we’ve been providing translation services to market research businesses for more than 20 years. In that time, we’ve learned what it takes to give organisations the support they need to thrive in international markets and understand their consumers, no matter where they are. So, if you’d like to learn more about how we can support your business with our specialist market research services, feel free to get in touch.