Marketing research one size does not fit all
Marketing research is pivotal in understanding consumer behavior, identifying market trends, and informing strategic decision-making.
However, the traditional approach of using open-ended questions to gather insights from respondents has its drawbacks. One of the main challenges is that respondents often don’t know why they do what they do, making it difficult to extract actionable information from their answers.
We argue that open-ended questions can be problematic in quantitative marketing research and proposes an alternative method utilized by Stealing Share involving hypothesis formulation and the use of 5-point Likert scales.
The Limitations of Open-Ended Questions
Open-ended questions allow respondents to freely express their thoughts, opinions, and experiences without any predefined response options. While this approach may seem valuable in capturing rich, unfiltered data, it presents challenges in quantitative marketing research. One significant limitation is the difficulty in extracting meaningful insights from the responses.
- Subjectivity and Ambiguity
Respondents often provide subjective, ambiguous, or incomplete answers when faced with open-ended questions. They might be unable to articulate the underlying motivations or the subconscious influences driving their choices. As a result, researchers may struggle to identify patterns or trends to guide strategic decision-making. - Memory and Recall Bias
Another issue with open-ended questions is the reliance on respondents’ memory and recall abilities. Human memory is inherently fallible, and individuals may not accurately recall past experiences or factors influencing their decisions. This can lead to misleading or inaccurate information, undermining the reliability of the research findings. - Response Bias
Open-ended questions may also suffer from response bias, where respondents provide answers, they believe the researcher wants to hear or what they perceive as socially desirable. This phenomenon compromises the objectivity of the data and reduces the validity of the research outcomes.
The Alternative Marketing Research Approach: Hypothesis Formulation and Likert Scales
There are many types of marketing research.
Stealing Share adopts a different methodology to address the challenges associated with open-ended questions. 
Instead of relying solely on free-form responses, they use hypothesis formulation and 5-point Likert scales for data collection.
- Hypothesis Formulation
Before conducting market research, Stealing Share spends weeks formulating hypotheses based on market understanding, consumer behavior, and relevant data. We use our behavior modeling to suggest hypotheses. These hypotheses are educated guesses about consumer preferences and the factors influencing decision-making. Formulating hypotheses helps focus the research effort, ensuring the study’s objectives align with specific business questions. Additionally, this approach empowers researchers to design surveys and experiments targeting the areas of interest, providing more precise and actionable results. - Likert Scales for Quantitative Data
To test the formulated hypotheses, Stealing Share employs 5-point Likert scales to collect quantitative data. These scales allow respondents to rate their agreement or disagreement with a series of statements related to the research topic. The Likert scale responses are then quantifiable, making it easier to analyze and interpret the data.
Using Likert scales for marketing research provides several benefits over open-ended questions
Standardized Responses
Likert scales offer predefined response options, ensuring consistency in the data collected. This allows for more accurate comparisons between different respondents and research studies.
Simplified Analysis
The numerical nature of Likert scale responses facilitates straightforward data analysis, enabling researchers to identify trends and draw meaningful conclusions efficiently.
Reduced Ambiguity
By providing respondents with specific response categories, Likert scales minimize the ambiguity often encountered in open-ended questions, leading to clearer and more precise insights.
Ease of Responding
Respondents find it easier to answer Likert scale questions easier and less time-consuming than crafting thoughtful responses to open-ended questions.
Advantages of the Stealing Share Approach
By utilizing hypothesis formulation and 5-point Likert scales, Stealing Share optimizes our marketing research process and gains several advantages:
Data-Driven Decision Making
Using Likert scales generates quantifiable data that is easily analyzable, facilitating data-driven decision-making for clients. This approach enhances the reliability and objectivity of the research outcomes.
Efficient Resource Allocation
Stealing Share allocates quantitative marketing research resources effectively by focusing on hypotheses, saving time and costs associated with sifting through vast amounts of open-ended responses. This efficiency allows them to deliver more value to their clients.
Improved Research Validity
Our quantitative marketing research process ensures that the study addresses specific business questions and objectives. This approach increases the research’s validity and the relevance of the findings to real-world scenarios.
Marketing Research background. CATI
The target market is more accessible to reach than in the past. And technology allowed more respondents to ignore the call.
When we started Resultant Research, our sister marketing research company, the CATI system was the gold standard of any market study.
CATI stands for Computer Assisted Telephone Interview. Don’t confuse this with automated robotic calls that drive us all insane. A real person did the interviews.
The computer had a database of possible respondents based on the study criteria.
This could include demographics, product usage, educational levels, income, and everything you need to fine-tune your audience.
The CATI nuts and bolts
The computer then randomized the respondent’s phone numbers and dialed them individually.
If the respondent answered the phone, the interviewer would explain the study and offer the respondent an honorarium to finish the marketing research survey. If agreed, the call will last about 15-20 minutes.
The computer showed the interviewer the question to ask, one at a time, and in some cases, randomized those questions to limit bias.
CATI marketing research relied on its methodology to be projectable. They were randomized and double-blinded (the researcher asking the questions did not know who commissioned the study).
CATI marketing research is expensive to field
Calling home phone numbers or business numbers was tedious.
And an element known as “incidence” weighs heavily. In shorthand, the incidence was the relationship between the number of people spoken with and the percentage of those that completed the study.
Let’s say, for example, the incidence level was 10%. This could be an indication of bias. Why did 90% refuse to participate?
Maybe they were busier people? Perhaps they were bored and wanted just to talk? You pay an enormous incentive to the person taking the studies to keep incidence high.
In the marketing research work, we did in medical devices (we conducted research on six continents for Medtronic, St Jude Medical (now Abbot), Wright Medical, and Integra Life Sciences and Biomimetic), speaking with an interventional cardiologist, surgeons and others cost $500 an interview. With an N of 500, these studies were uber-expensive.
For a short while, adding cell phone numbers improved the incidence. But today, few will answer a mobile call if they do not know the caller’s ID.
CATI studies are rare today. It became hideously expensive, and the data was not as reliable as in the past because of low incidence and bias.
The rise of Expert Panel Marketing Research Studies
Expert Panel marketing research studies have replaced CATI as the gold standard of market research.
They remain highly cost-effective and have become as projectable as the CATI marketing research of old.
These are online studies, but they are not EMAIL studies. Email studies are not projectable or reliable.
They have a bias from each end of a standard bell curve. Those who love or hate you are the only people willing to take an email study.
The rest can’t be bothered.
So, we don’t do email market studies. They are dirt cheap and as valuable as dirt.
An Expert Panel Study (sometimes simply called a Panel Study) is a different marketing research tool. But they rely on their value based on the representation of the panel that has been recruited.
Marketing Research today
Research Companies like Resultant Research (our in-house research company) partner with these professional recruiters who compile the panel.
It works something like this. Possible respondents are called and asked if, for a fee, they would be willing to participate in marketing research studies. A fee is negotiated, and all their demographics are recorded and entered into a database.
So, if your marketing research study is, for example, Heads of Households with natural gas heat, the database is scrubbed for those demographics.
If you are part of the panel, the system randomizes the possible respondents and sends them a link to take the study.
If it is a representative panel, you get excellent results for less money and in a shorter time.
These studies also can share visuals or audio, something impossible with CATI studies.
We started testing panel studies alongside CATI studies years ago. We found identical results from both market research studies as the panels improved.
Call us—one caveat. Any marketing research study is only as good as the questions asked. A Stealing Share, we ask better questions. It is that simple.
We spend more time creating the survey questions than fielding the study itself. That’s why we are the go-to choice for Fortune 500 companies across the globe.
Let us help you uncover the insights in your market that steal market share.
