Ecommerce product research – Why focus groups don’t work & what to do instead



I found myself in Melbourne, Australia, sitting behind a one-way glass, observing a group of 12 people gathered around a wooden table, scrutinising a metal drink can. This metal object was a prototype poised to disrupt the wine-in-a-can market (and if this one didn’t, then perhaps one of the other ten would).

This was an ecommerce product research focus group. At the time, I was employed in the innovation department of a large Australian company, tasked with developing new product concepts for our major beverage brand clients. I was pretty excited to be there and believed the concepts being put forward were great (I still do).

focus group

However, the setting was unusual. Considering that this product would be bought in a store or online, the environment we found ourselves in was more akin to a school classroom, complete with a teacher (the focus group moderator).

Alright, so the atmosphere didn’t mirror the future purchasing context of consumers. How about the interaction between the moderator and the participants? That was far from perfect as well. The moderator would present a prototype, provide some minimal context, then ask for initial impressions and some follow-up questions. Participants were hesitant to share their thoughts, often leading to one or two dominant voices in the room, quickly echoed by the rest. Disagreeing with others isn’t easy, and being in a focus group doesn’t change that.

Crowd psychology indicates that group dynamics can differ significantly from the actions and preferences of individuals when they are alone. Crowd psychology – Wikipedia

We had some female-oriented concepts (Rosé in a can), some male-oriented products (Shiraz in a can), and some that didn’t fall into either category. Acquiring representative data for of the entire market and its demographic makeup from the ten participants behind the one-way glass seemed a bit of a stretch.

That’s one way to spend $4,000 to $12,000 on ecommerce product research if you’re a larger company, and who am I to judge? Nonetheless, the experience has stayed with me.

I’m now going to outline the pros and cons of focus groups as they relate to ecommerce brands (from my perspective), then suggest a free alternative, followed by my vision of the ideal solution for this ‘new product research’ area (which, you guessed it, we have an app coming out for soon).

Why choose focus groups for ecommerce product research?

Let’s examine a few reasons why you might decide to run a focus group for new product development/research.

  • Speaking with customers is generally regarded as a great idea in ecommerce product research, SaaS, and start-ups. Although when your sample size is a 10 person focus group, it can be misleading. For example, insights from 10% of 10 people shouldn’t carry the same weight as 10% of 1000 people on your website.
  • A little bit pessimistic, but from an executive’s perspective at a larger company, being able to shrug your shoulders and say “the focus group liked it” can be a handy escape route.

The illusive perfect product

What to do instead?

One method I’ve encountered a few times is engaging an online community of your customers, primarily Facebook groups.

At first glance, it would still be subject to group dynamics that don’t represent the typical online store purchasing experience. However, I believe there’s a reduced ‘voices dominating the room’ effect. Also, it’s not the same user situation as your online storefront, but it is digital (which is much closer than focus groups).

The main advantage of this approach is that it’s almost guaranteed to involve your actual customers. These individuals appreciate and support your brand, spending real money on your products. Determining what these people like and how to create offerings that are a perfect fit for them is far more valuable than receiving feedback from a randomly sourced audience answering for a hypothetical situation.

Ranking products

A new approach for ecommerce product research

On a day-to-day basis, I manage a pre-order solution for Shopify stores called PreProduct. We primarily work with stores launching new products. I am passionate about the space, and the business is going well.

That said, I’ve held this opinion on focus groups for years. So, I couldn’t resist taking a shot at solving the problem. It’s also somewhat related to PreProduct, but focuses on the beginning of the new product cycle instead of the mid to end.

What would the perfect solution be and do?

  1. Something that takes the advantages of the Facebook Groups approach, but is even more store-native and representative of the buyer journey. It should:

    … Reside on your storefront, i.e. Ask the people who matter, when they’re in the setting that matters… your website.

  2. The Facebook groups approach is more representative of your customer base than a focus group, but it excludes a significant portion of your audience. You have to be into Facebook groups and be the type of person who has the time and inclination to hang out there.

    … Therefore, polling on your storefront makes sense, but the process of evaluating new products needs to be quick and convenient. If we want additional information from them, that’s fine, but let’s ask them that later. Getting a primary vote/score should be prioritised.

  3. The actual voting/ranking action needs to be as similar to a buying action as possible (without causing confusion).

    … We decided on a design that borrows visual cues from a collection page’s grid item. With the product image occupying a significant portion of the space, alongside a product title and an option to show benefits and descriptions.

  4. I don’t think people buy aspirational/lifestyle products that are a 7 out of 10, as such, we need some binary options here. Is it a ‘yes’ or a pass? Do you love it? If you love it, do you want us to email you when it’s available? We want to figure out if people will buy the product, so won’t be offering a typical sliding scale answer approach. Points on a scale can mean different things to different people.

    … There’s a digital online format that already does an excellent job with binary answers: dating apps. We’ve implemented a swipe-left/swipe-right interface, along with buttons for pass, no, love, and an email capture on the end screen.

  5. Lastly, it needs to be affordable and quick for merchants to create these product ‘pitches’ and start gathering votes, e.g. uploading a name and an image should be enough. We’re also implementing a way to easily use AI to generate new variant images via prompting, but that can wait for another time.


In conclusion, I hope the above has made you reconsider some of your assumptions about running ecommerce product research. If you think our approach sounds compelling, I’d love you to sign up for our early-access list below (the new app will be called Product Love)