Generative and evaluative research in experience design

By: Sean Smith
Date: March, 2017
File under: Articles

At U1 Group, we are continually working with our clients to assist them in making informed decisions about the experience design of their products and services, from the initial discovery phase right through to delivery. With the progressive maturation of the UX industry we have observed an increased appreciation and understanding of the value delivered by user and customer experience research.

One of the biggest challenges we see individuals and teams grappling with is understanding how they can make the best use of research to make experience design decisions. Particularly in regard to the best time to engage in research and the type of research they should be conducting.

We want to share our experience on this topic to hopefully assist in making decisions about the type of user and customer research to engage and when, whether you are conducting the research yourself or engaging a third party.


Research definitions

Let’s start by defining two types of research most relevant to the experience design of products and services, generative and evaluative.

Generative research

Generative research should occur at the beginning of the product or service design journey. This discovery phase is where generative research techniques can assist you in defining the product or service and leads to a strong understanding of your intended audience and their relationship to your organisation/brand and the product or service you intend to deliver.

As opposed to market research which explores potential markets to push into, generative research is exploratory in nature and pulls from participants in order to inspire the design of a new product or solution. With generative research, you are not always sure what the answers or outcomes will be. However, as the insights typically identify the needs and motivations of your intended audience, they provide a strong foundation for making experience design decisions.

With generative research you are pulling ideas to inspire or inform the design of a new product or solution

Evaluative research

Evaluative research should occur once the experience design has taken shape in some form. That is, there is something that can actually be evaluated by your intended audience. The fidelity of this experience can vary greatly, within a digital context this could be from some sketch drawings of an app through to a live version of an app that is already in market.

Outside of the digital context it might be a journey map outlining the intended touch points of a planned service through to direct observation of customers’ interactions (online and offline) with an existing service or product. In a nutshell, you are seeking feedback on something that already exists, whether it is in the form of a prototype or “live”.  Through conducting evaluative research are you seeking insight that will help to refine and improve your experience design.

With evaluative research you are seeking feedback to refine or improve an existing idea or design


Generative or evaluative research?

So, broadly speaking we can say that generative research applies to the definition of a product or service whereas evaluative research applies to its refinement and improvement. The time to apply generative research is before you have committed to the experience design, whereas evaluative research is best used once the experience design has taken form. Generative research tends to be one-off and in-depth, whereas evaluative research should be applied regularly through the design, development and delivery of a product or service and can be light touch if required.

Where we see people getting into difficulty is where due to a lack of understanding or budget constraints they attempt to combine both types of research. Evaluative research should be applied when you understand the problem you are trying to address and/or the experience you want to create and are striving to achieve the best implementation. Evaluative research will not help you to build the “right” product or service, but it will help you to ensure what you build is built “right”, or whether you are actually building the “right” thing.

You should avoid thinking in terms of generative versus evaluative research, as they are equally important in delivering the best possible products and services, but rather when to engage in each type of research. It is also important to recognise that there are a number of research techniques and tools that are common to both types of research. Qualitative and quantitative research can be used for both research types, something we will address in a future post.

Back to articles
Share this

Get in touch

Find us at:

Level 24, 570 Bourke Street,
Melbourne VIC 3000
(03) 9684 3470


Online enquiry