Online shopping saw an unprecedented rise in popularity in 2020. In new research published by Google, the authors of Decoding Decisions: The Messy Middle of Purchase Behavior propose an updated model for understanding how consumer purchasing decisions are influenced and made. They argue that 'sitting between twin poles of trigger and purchase' is what is called the "Messy Middle." This is "a space of abundant information and limitless choice that shoppers have learned to manage using a range of cognitive shortcuts".

The researchers engaged in a two-year project to learn how consumers use the information and choices provided through the internet to make online purchases. Using insights gleaned from Google Trends, Behavioural Science Literature, and conducting simulations with actual consumers, this is Google's latest attempt to answer the holy grail of marketing questions - How do consumers decide what to buy and who to buy it from?

It builds on more than a century of work including Elmo Lewis' original AIDA model, The McKinsey Consumer Decision Journey, Jan Carlzon's Moment of Truth, Ehrenberg's Post-purchase experience, Google's own Zero Moment of Truth (ZMOT), and others.  While not entirely new, the study reflects the complex nature of how people make decisions especially in the age of the Internet. And adds to a growing body of research on cognitive bias and consumer behavior. It also provides some important updates to our current understanding of the customer purchase journey.

In this article I examine some of the key insight from the research, and offer some tactical suggestions on how firms can incorporate them into their campaigns in 2021, and beyond.

But first, here are some key highlights:

• 'Online consumers don't tread a typical journey. Instead, there is a confusing web of touchpoints spanning purchase journeys'. The authors break the model down into 4 E's - Exposure, Exploration, Evaluation, and Experience.

‍

• Exposure is the aggregate of everything you've ever seen or heard about a product or category. Every ad watched or clicked, testimonial read, word-of-mouth heard and information seen, learnt, or read in the press or anywhere online or offline.
• 'Before making a purchase decision, consumers explore their options and expand their knowledge and consideration sets, then evaluate the options and narrow down their choices. This process translates into two distinct mental modes: exploration and evaluation, where 'the first is expansive and the second is reductive.'
• Experience, sits outside of the "messy middle"- and speaks to the experience a customer has with the product or service they've purchased. It feeds directly into their background exposure. And may manifest as online reviews, or other feedback. If the experience is especially bad. customers may end up leaving the category altogether.

But here's where things get interesting:

• The research concludes that consumers tend to look for cognitive shortcuts when making buying decisions. Though rational and emotional elements figure into making purchases like the complexity of the purchase, the price of the item, or emotional triggers such as nostalgia, there are powerful cognitive biases at play as well.
• There are dozens of such biases which exist. And generally, there are mental models which we use to make sense of the world around us, and to make decisions. But the authors propose that including any or all of the following six biases will prod consumers to make a choice in your favor. You've probably seen or used them before:

‍

‍

1. Category Heuristics: Consumers use rules of thumb to make quick, reliable decisions. These are known and accepted characteristics about a product or category. Like Miles Per Gallon (or Miles per Charge), number of Megapixels, number of Carats in a diamond etc. The principle has to be simple to grasp, easy to access and be precise in its description.
2. Authority Bias - If someone knowledgeable about the category or product says it is so, then you'll tend to believe it. Incidentally, the research revealed that independent reviewers were considered more trustworthy than industry publications (yeah YouTubers!)
3. Social Proof - People follow people. If other people are using it, and say it's good then it must be good - apparently.
4. Power of Now - People are inherently impatient. Instant Access, Same Day Shipping, 24-hour delivery, Ready in 5 minutes, or any variation of these are as good as gold.
5. Scarcity Bias - Available for a Limited Time, Offer Ends Today, Only 10 Left. Ladies Only. Any permutation of these will hook your buyer, they argue.
6. Power of Free: 'The power of free is a source of irrational excitement, and can be critical in persuading a consumer to buy'. Buy one get one Free, Free Gift Card with every purchase etc.

How did they arrive at this conclusion?

'The researchers selected "31 products representing a broad range of risk, complexity, and emotional and financial investment, covering several major verticals and sectors, including travel, financial services, consumer packaged goods, retail, and utilities"

• They surveyed 1,000 people for each category - a total of 31,000 participants in all (n=31,000). Each participant had to be familiar with online shopping, and in the market for that particular product. They excluded anyone who had already made up their mind.
• Participants were given a total 10 purchase simulations, 310,000 purchase scenarios in all - and asked to choose their first and second choice of brand in the respective category. From this, the team was able to test how the six biases configured in their purchase. To control the prejudicial effect against the presence or absence of the same bias, they paired different levels of execution - (Same-day shipping etc.)

• They created a very basic presentation - no creative bells and whistles - just two logos with all relevant information placed side by side. The participants were then asked to choose their purchase preference using the same process they would follow in real life.
• The team was able to visualize and measure the impact of how changing or adding any single element, or combination of elements resulted in an increased or decreased share of consumer preference for the respective brand.
• The consumer path to purchase they observed looked something like this:

Between the trigger at the top-end, and the purchase at the bottom-end, consumers take a confusing maze of routes to their purchase, these include "search engines, review sites, video sharing sites, portals, social media, comparison sites, forums, interest groups/clubs, retailer sites, aggregators, blogging sites, voucher/coupon sites, branded sites, publishers, and noticeboards" - this is the messy middle.

...the following insights also emerged:

• Presence Produces Preference: The researchers found that "simply giving the shopper the option to choose their second choice brand was enough to entice 30% away from their initial choice". The idea seems pretty self-evident, but by simply being there when your customer needs to make a decision you increase your chances of purchase.
• Perceived Value Beats Loyalty In Most Cases- The second choice brand [was] able to take a full 90% of preference away from the first choice brand when supercharged with all six biases." The greater the perceived value, the better the chance of purchase.
• Never Underestimate the Newcomer - Across all 31 categories surveyed, customers when fed the right incentives (biases) were at varying levels willing to switch from their first choice to a fictional brand to which they had no prior exposure.

• Brands matter, still - Though some consumers were willing to switch to a brand they never heard of before, and to which they had no prior exposure, many were still willing to stick to their first choice brand even when their brand had a vastly inferior offer.

The limitations of the "Messy Middle" are many, but the implications are far-reaching:

‍

The research did cover some ground in terms of the universe of online purchases. But didn't cover SaaS products, B2B Products, or Professional Services. However, one can still draw some helpful inferences. For example, the need to ensure product information is abundant, easy to find, and easy to understand. And being in the consumer's consideration set - whether through retargeting ads, social media, or the press is crucial to them deciding in your favor. But beyond these limitations, the research has some far-reaching implications:

On the Bright Side:

• The research opens up a broad spectrum of opportunities for marketers and researchers alike to explore how different mental models, modes, personality types, and other aspects of behavioral science can be used to understand online consumers, and build delightful and engaging experiences around their preferences.
• These insights also make it possible for almost anyone to do well online. In the fashion space for example, entrepreneurs now have ample opportunities using platforms like CA.LA to create new brands, and dominate entire categories using the power of social media to amplify exposure, and leveraging these biases to grab market share from established players. Shopify dropshippers make a killing using these tactics.
• For established brands, simply tweaking existing offers through better copy and design around these biases, and consistent with what customers value, can go a far way in driving revenue without the need for a brand overhaul or repositioning.
• Brands can also align these insights with their internal and competitive data, content strategy, and automation tools to better amplify exposure, and create an environment where consumers have fewer options to consider, thus making the path from trigger to purchase less convoluted, and more in their favor.

On the Dark side:

• The marriage of behavioural science and predictive algorithms present an ethical dilemma when taken to the limit. The ability to understand consumer behavior at a granular, even personal level, and being able to target offers so precisely based on these biases, means some brands can not only predict consumer behavior, but control it.
• Companies with big datasets and better AI models can use these biases to tip the scales of Capitalism in their favor making it near impossible for other players to compete.
• Fake news, fake reviews, fake fans, and fake influencers aided by automated bots and AI algorithms, and loaded with these biases can produce an unfair advantage for bad actors that can threaten the existence of some categories, and make the shopping experience undesirable overall.
• Across some verticals, Mortgage and Business Loans for example, leveraging these biases at scale can reinforce historical discrimination against certain social groups.

How to Leverage Cognitive Bias To Make More Sales in 2021:

• Answer Questions: The messy middle provides a range of new opportunities for brands to make the path to purchase less complicated. This includes building SEO strategy around question-answering. The idea here is that consumers explore and evaluate their options by asking questions. Free tools like Answer Socrates can reveal top trending questions being asked on search engines. Find the questions being asked in your category, and build content around those answers. Question-answering bots can transform your website into a one-stop destination for consumers to explore and evaluate choices in your category - without them leaving your website.
• Leverage Competitive Data to Inform Creative Decisions: This is as old as marketing itself, but there is a growing list of tools that use artificial intelligence to identify patterns in social media content in your category. Knowing what your competitors' fans, followers, and customers like and engage with on social media can give you a distinct advantage when coming up with campaigns of your own. You can learn more about the technology here, or download a case study here.
• Retargeting Ads Are Your Best Friend: Retargeting ads are basically recurring advertising pitches. They use cookies to track consumers more dynamically and deeply, and target messaging wherever they are looking for your product. They focus on prospects who have already visited your website, and might be in the market to buy.
• Automate Campaign Testing: Knowing which offers work best can be a time-consuming activity. Thankfully, there are machine learning models that can help you fast track your A/B and multivariate testing, and help you manage your creative allocation and digital ad budgets for a better return on spend.
• Fast, Mobile, and Frictionless: Regardless of the category in which you operate, creating an experience that is easy-to-navigate can give you a significant advantage. If the "messy middle" is all about exploration and evaluation, then the extent to which your site is easy to access and understand will go a long way in making your case.
• Content remains King: The challenge of producing helpful, relevant, well-produced content on a consistent basis is difficult. It's doubly hard if you're operating in a new category where not much has been written or said about your product. And just as difficult is taking on an established competitor with years worth of content. But inside the messy middle, content is still the best investment you can make in your brand for the long term. Today there are a raft and AI-powered tools and services that make creating great content a little easier

Leveraging these tactics across your various touchpoints, and building variations of the six biases into your campaign offers are a sure-fire way to propel your sales in 2021.

***

Please note: All figures and images cited have been extracted from the report: See full document.