News

A major new study into the online behaviour of Amazon users

Leave a Comment

Amazon Box Robot
Amazon is a giant in online product search and purchase, with its influence and category reach growing all the time. The company’s latest move is a bold incursion into the grocery sector with its acquisition of Wholefoods and the recent launch of the Amazon Go convenience store format.

Few people doubt that Amazon’s C-Suite crunched a lot of their own platform data before making these moves. And even fewer doubt Amazon’s ability to wring every bit of value from the data they collect as the company’s attack on new categories win sales. Amazon carefully guards its proprietary customer data, and with good reason. It poses a direct threat to any established retail and consumer brand that cannot leverage the same degree of data dominance to make decisions and defend its market.

Some of this data is made available to big brands or vendors selling on Amazon’s platform through a program called Amazon Retail Analytics Premium. But it is expensive, with vendors paying 1 percent of their wholesale cost of goods sold to Amazon or a minimum of $100,000 to get access to a database that lets them see some, but not all, of the data Amazon has compiled.

If other businesses could get similar insights on consumers’ use, and importantly, non-use of Amazon for products in their category it would begin to significantly level the retail playing field. This is the inspiration behind our major new research study into the online behaviour, shopping habits and digital lifestyle of Amazon users.

Colourtext and Netquest have teamed up to collect and analyse every click made by more than 3,000 U.S. Amazon users on their mobile and desktop devices. Colourtext has leveraged the power of Netquest’s U.S. behavioural data consumer panel to reveal how users and non-users of Amazon use the online giant’s platform to search for and complete retail purchases. This includes a detailed segmentation analysis to uncover important differences in behaviour and attitude between Amazon Loyalists and Amazon Avoiders.

The click-stream analysis has been supplemented by data from a follow-up online survey of 2,279 Amazon users who contributed their behavioural data to our study. This yielded a wealth of textured attitudinal and perception behaviour that’s played a key role in interpreting and articulating the shopping patterns identified in device usage data.

Using this data we can identify the consumers who are most likely to abandon their shopping cart, or indeed complete a purchase, with details on their age, gender, income level and online content consumption profile. Retailers and brand owners in any category will want to use these insights to plan their own strategy for making Amazon work in their favour. The insights from this major study can also help businesses figure out how to build a moat around themselves and fend of future category incursions from Amazon.

You can view a summary slide deck of this study here.

To learn more about the Amazon user data we have collected, or data we have on other categories and brands in the US, UK, France, Germany, Italy or Spain, you can reach us by email.

Want to break into a new market discovered by a competitor?

Leave a Comment

beau-runsten-466663-unsplash-2

We’re learning how to find and target another brand’s early adopters.

My client is launching a product into a new category but they’re not first into this market. A competitor has recently established itself and generated big sales, which is great because it proves customer demand actually exists.

We’re therefore more than curious to discover who the early adopters are that buy our competitor’s brand. But how can we do this when information is thin on the ground?

Here’s something we tried that seems to be working. Our competitor has around 10,000 followers on Twitter. We used Twitter’s free Public API to drawn down the user profiles for each of our competitor’s followers. This data usually contains a short personal description that people write about themselves in natural language.

This profile data is incredibly rich but can be difficult to analyse at scale because it’s totally free-form and unstructured. However, there are a bunch of Natural Language Analysis techniques we can use to make this stuff useful.

We began by running all of the personal descriptions we got from Twitter’s API through a semantic tagging app. This classified all of the different things people said about themselves into meaningful semantic categories such as Family, Information Technology, Science, Education, Art and Music etc.

After a little manual refinement we developed 40 bespoke semantic categories that were statistically significant within the personal descriptions of the early adopters who follow (and presumably buy) our competitor’s product (see Chart 1).

Screen Shot 2018-08-17 at 10.43.24

This information begins to paint a useful picture of the early adopters we’re trying to reach and understand, but we can take the analysis further. We used a graph-based segmentation analysis to group individuals into clusters based on the semantic tags that appeared in their self-descriptions. This yielded 9 early adopter customer segments with rich personality characteristics that we can target with our brand launch proposition (see Chart 2).

Screen Shot 2018-08-17 at 10.43.28

The largest group in this emerging market sector (16%) are “Creative & arty” types. A typical self description for this group reads, “Graphic designer, love my job, photography, nature trails, waterfalls, the coast and LAUGHTER”.

My personal favourites are the “Ironman Entrepreneurs” – this group seem so clear to me. 80% are men, 8% live in the SF Bay Area and a typical self-description goes like this, “Entrepreneur interested in technology, innovation, extreme fitness, free expression, sustainable development, exponential results, and wide-scale collaboration”.

Now I feel more confident about who these early adopters are and what they need to hear from us to consider switching brands. All thanks to a bit of free data from Twitter!

Colourtext Alerts

Leave a Comment

http://www.colourtext.com/wp-content/uploads/2018/08/Sale.png

What are Colourtext Alerts?

Colourtext Alerts are a simple, low-cost way for media companies to drive revenue growth from local and regional advertisers.

It’s a simple online service that delivers media sales leads, once a day, directly to an exec’s inbox or Salesforce account when local companies use social media to promote new products, services, launches and events.

People who want to monitor retail competitors or general commercial activity in any local market also find the service useful.

Why use Colourtext Alerts?

Media sales teams make hundreds of cold calls every day that often begin something like this, “Are you thinking of doing any advertising at present?”

Ninety-five out of one hundred calls are wasted because the prospect isn’t ready to advertise. In contrast, Colourtext Alerts are automatically generated when companies actively push a new proposition on social media that they want to promote.

This means a huge leap in sales process efficiency is gained when an exec follows up on a Colourtext Alert – they will always be talking to a customer who is currently thinking pretty hard about advertising!

http://www.colourtext.com/wp-content/themes/starkers-master/img/img_section_2.jpg

How does it work?

Colourtext Alerts uses machine learning technology to sift through social media and spot the moments when local companies are in-market with live offers that could benefit from press, radio or outdoor advertising. The system has been delivering hundreds of actionable leads every day for more than four years to teams throughout the UK.

What does it cost?

It costs only £200+VAT per month to provide Alerts coverage of a town or city area for everyone in your sales team. Generating sales leads for a bigger region like the Northwest or Yorkshire is just £300+VAT per month. Our minimum sign-up period is 3 months and we only need 1 month’s notice to close your account.

Colourtext Alerts are a simple, low-cost service delivering a big ROI for sales teams that want a more efficient way to find revenue opportunities in their local market.

Start a free trial

Colourtext Alerts is simple, easy and great value but we’d like you to try it for free before you invest. All we need to get you started is a contact list for your team members (so they can begin receiving their sales leads) and the main postcode zones of your local or regional market. Send a quick note by email to discuss how we can set up a free 4 week trial for your team with the minimum of fuss or delay.

Fighting back against Amazon

Leave a Comment

clark-street-mercantile-33915-unsplash 570

Fighting back against Amazon – segmenting online shoppers in the US

Market segmentation has been a nerdy personal passion for me since I worked on consumer magazines in the 90s and 00s. I got to create projects for Emap that explored new trends in music, luxury branding, technology and popular culture, which was ace. We even gave some of those studies names like Super Youth, ROAR and Project Phoenix.

Since that time Colourtext has built upon this experience by developing segmentation methodologies that analyse new digital data sets and can operate at global scale. One market of particular interest right now is Retail. This is a sector where brands that depend upon traditional physical store visits to drive their trade increasingly feel the pinch, especially from Amazon.

To learn more about this massive sea change for both consumers and traditional retail brands, Colourtext has teamed up with Netquest, the online research panel company. We used Netquest’s US consumer panel to investigate how online shopping behaviour is evolving in the US.

We did a market segmentation study to reveal e-Shopper behaviours amongst a range of familiar and newly emergent consumer groups. Moreover, we paid particular attention to how different segments are using Amazon. We think the research results provide insights on how retailers can not only survive but also thrive in an environment where online competitors like Amazon increasingly shape the weather.

You can read and download a copy of our summary report here.

Best practice in survey design for segmentation

Leave a Comment

Admap cover edit

Segmentation research is an effective tool for defining your target audience, aided by an abundance of data generated from digital platforms and mobile devices. But the secret of success lies in good segmentation survey design.

We wrote an article for Admap with our friends Jon Puleston and Alex Wheatley at Lightspeed Research. It looks at best practices to design a segmentation that’s sensitive to consumer and cultural nuance, capable of operating at scale and immunised against the danger of survivor bias. Read it here if you don’t have an Admap subscription.

If you are considering a new segmentation strategy that aims to transform your organisation why not get in touch by by email. It’s free to talk and there’s a chance we’ve got ideas that can super-charge your plans.