How to pick a market research approach?

Image of woman and brain

To read the Japanese version of this post (from Mr Ryota Sano) click here.

For most business needs there are several good ways to find a market research approach. The table below sets out some thoughts about the most widely used techniques for four categories of business needs.

1 How many X do Y?
Within this heading I am including product/service usage and attitude mapping, customer satisfaction, and ad tracking and awareness. The key need is to quantify things so that they can be managed. For example, quantitative ad tracking allows the buyers of advertising to measure the effectiveness of their expenditure, at least in terms of the number of people reached, the number recalling advertising, and measures such as brand recall and stated purchase intention.

There are two key MR techniques:

  • Surveys
  • Passive / Big Data (including social media)

2 Reactions To & Predictions About a New Y or New Marketing for Y?
This heading relates to testing new products, new advertising, new marketing etc. The needs range from understanding how people interpret and react to a new product through to predictions about future sales.

There are a wide range of MR techniques available for this category of needs:

  • Surveys (plus modelling for predictions)
  • Implicit and non-conscious techniques
  • Usability testing
  • A/B testing (for predictions, not understanding)
  • Prediction markets
  • Qualitative research (for understanding, not predictions)

3 Why do people do Y?
In most cases, companies want to understand why people do things because they want to identify options for change, options for improvements, options for new uses or products.

There is widespread agreement that you can’t simply ask people why they do things, the key MR options are:

  • Qualitative research
  • Surveys with advanced analysis (e.g. regression or conjoint), possibly utilising implicit and non-conscious techniques.
  • Behavioural Economics – e.g. setting up a controlled experiment

4 New Ideas for Y
This is an area of growing interest, accompanied by the realisation that there are more smart people outside your company than inside it.

Banner for Festival of NewMRThe two main MR approaches for generating new ideas for a brand or service are:

  • Qualitative research
  • Crowdsourcing platforms (which usually combine suggestions, edits, and some form of voting or prediction market assessment).

Do you feel my four headings cover most of the key needs that businesses have? What about the techniques, are there others that are in common use?

Note, you may feel that one or more of these techniques is not suitable (e.g. surveys for understanding why), but my point is about which techniques are in broad/common usage.

Japanese version contributed by Mr Ryota Sano, from TALKEYE Inc., Japan.



1) どれだけの数(X)がY(対象)をしているのか?


  • 質問紙調査
  • パッシブ/ビッグデータ(ソーシャルメディアを含む)


2) 新規なYまたはYに対する新しいマーケティング手法についての生活者の反応および予測


  • 質問紙調査(予測モデルも)
  • 暗黙的および無意識を利用した手法
  • ユーザビリティテスト
  • A/Bテスト(注:予測はできるが、理解はできない)
  • 予測市場
  • 質的調査(注:理解はできるが、予測はできない)


3) 人々がなぜYをするのか?


  • 質的調査
  • 高度な分析手法を用いた質問紙調査(例:コンジョイント分析、回帰分析)、暗黙的および無意識を利用した手法も考慮
  • 行動経済学(例:統制環境下での実験)


4) Yについての新しいアイディア



  • 質的調査
  • クラウドソーシングのプラットフォーム(メンバーからのアイディア投稿、掲示版でのやりとり、投票機能、予測市場などの統合プラットフォーム)





5 thoughts on “How to pick a market research approach?

  1. This is actually really useful! I am curious, where (or does) customer experience (e.g. journey mapping, ethnography) fit in here? It’s a bit of New Ideas for Y and Why do People do Y, I think. Fits under qualitative maybe.

  2. Hi Laura, many thanks for the comment. When customer experience is quant, it is typically How many X do Y, including defining the volumes associated with different paths to purchase. If we add some smart analytics to that quant we might identify, that people change brand because they have a bad experience, or because they experience effective point of sale material, so that would be Why do people do Y.

    If we are talking about qual, for example ethnography, then we are typically looking at Why people do Y. When we understand Why people do Y, we might hope to get some new ideas for Y.

  3. Hi Ray,

    Thanks for this useful simplification. It’s great for sharing with clients who are new to research.

    I would add a subtle distinction to your crowdsourcing discussion. In my view, crowdsourcing is one form of co-creation, but in crowdsourcing, the “crowd” (by nature of being a crowd) remains somewhat distant from produce or service developers. Co-creation, IMO, is a much more intimate, ongoing process where a careful selected set consumers work closely with developers to conceptualize, create, and refine a new service or product. I would put crowdsourcing as a distinct, more distant form of co-creation.

Comments are closed.