How to never waste an ad again (thanks to iterative research)

Iterative testing is a fundamental cornerstone in product management and involves continually improving a product or service as it is released, tested, consumer feedback is implemented, and released again. 

It’s a well-known method in software development but can be applied to products, services and market research. 

In this article, I’ll cover the benefits of how iteration can work for marketing, share some best practices, explain why it works well for consumer packaged goods (CPG) and quick service restaurant (QSR) brands and more. 

Industries like CPG and QSR move fast. The good news? Iterative testing works well in sectors with quick cycles. 

More conventional methods of market research that work in a more linear fashion are slower, meaning product updates and changes take much longer. The faster pace of iterative testing means that consumers get what they want, sooner. 

Why? When a product or service can be introduced with less risk, and brands understand what customers want at a deeper level, everyone wins. Go-to-market times are shortened, consumers are happy and profits improve.

Iterative market research consists of four steps that are repeated as often as needed: 

1. Develop a concept

2. Test it with consumers

3. Get feedback and learn

4. Refine the concept

Each iteration of the product or service addresses direct consumer feedback, making the next release incrementally better. This cascade of gradual changes results in a product that aligns better and better with consumer desires and expectations. 

It’s best to use a mixed methods approach to iterative market research and include both quantitative and qualitative tests. That way you have both measurable data and information about how consumers feel and their opinions. 

Examples of quantitative tests include: 

MaxDiff Analysis

A best-worst test that asks people to choose which is the best and which is the worst from a set of options. For example, a brand that offers two flavors of a candy and asks consumers which is best and which is worst is using a MaxDiff test. 

TURF Analysis

 A Total Unduplicated Reach and Frequency (TURF) analysis narrows options of combinations. Imagine you sell potato chips, and can only produce four flavors. You know that plain will be one, because that’s the biggest seller. How do you decide the other three? A TURF analysis helps you see how many people would be interested in your products overall. That’s your reach. It also shows you how many consumers would be interested in each of your flavors. That’s the frequency part. 

Understanding which combination of flavors helps you to reach the most consumers, is exactly what a TURF analysis can do. In many cases, TURF data can be drawn from other types of tests, such as a MaxDiff analysis. 

Monadic testing

A monadic test presents a product or concept in isolation, to avoid comparisons. This allows consumers to provide specific feedback to just the one thing, rather than how they feel about it in comparison to another product or concept. For example, you might learn how a consumer feels about a scented candle labeled “forest” instead of whether they like “forest” more than “cotton candy.” 

A sequential monadic test is when you test a series of products, one after the other, so the consumer might evaluate “forest,” then “cotton candy,” and maybe one or two others after that. It’s worth noting that there’s some risk of order bias with sequential monadic testing, because the order of the products presented can influence how consumers evaluate them. 

Examples of qualitative tests include: 

Focus groups

A focus group is a small group of people from a brand’s target market. Formats are varied, and the meetings can be conducted online or in person. People answer questions about a product or concept, usually how, what and why questions. 

Interviews

Although individual interviews can be time-consuming, they are a tried-and-true way to gather information from a target audience or persona. 

Observational

People may be observed in person, or via camera, but the idea is to see how they interact with a product or concept in a natural setting. Observations can help brands see how people shop, navigate through a store, react to a product or what they are drawn to. 

Pulse survey

A pulse survey is very short and conducted with the same group over time. It’s useful in iterative testing to see if the incremental changes are impacting how consumers feel. 

Many other research methods are available. For iterative market research, using a mix and testing often are the two key concepts. Iteration isn’t only for research, though. It can also be used to create more accurate and robust customer segments.

Customer segmentation isn’t a new idea, but traditional segmentation and iterative segmentation are significantly different. 

Traditional segmentation uses basic demographic data, such as ethnicity, age and some slightly more advanced analytics like a cluster analysis of a specific study or survey. Brands divide customers into groups to better understand their behavior, needs and responses so that they can create campaigns that connect emotionally. 

The segments and personas that result are often two-dimensional and become stale quickly. 

By monitoring and adapting to audience preferences and evolving tastes, brands can adjust campaigns, products or concepts as needed. Establishing a feedback loop, through surveys, forums, reviews and direct engagement so that customers have the opportunity to express their thoughts allows brands to reshape their audience personas and update segments. 

Creating iterative segments requires multiple steps, slowly layering on more information as it’s obtained. The result is a much more nuanced grouping of customers that reflects trends and changes. 

Since every step of iterative market research involves data collection, it’s probably not surprising to find that customer segmentation does, too. Happily, the same surveys and observations and other market research can be used to segment audiences and develop audience personas. 

Here’s a few key steps: 

1. Customers may be grouped by demographics, behavior or other pertinent information.

2. Once the segments are established the next step is to prioritize them. Which groups make more purchases? Which are more likely to be interested in a new product? The overall business goals direct the prioritization of the segments. 

3. Create detailed personas for each segment. Each persona should have a name, backstory, photo and as many details as possible. Personas should include distinct behavioral patterns, too, such as buying a certain type of candy on a certain day of the week, or how often they take lunch to work versus eating at a quick service restaurant for lunch. 

4. Finally, validating the personas means testing them. Brands may present scenarios for each persona to a focus group or an individual in an interview. 

5. Repeat!

The goal, as with all iterative testing, is to gather feedback, make adjustments and repeat the process so that the personas are continually updated, refined and become increasingly reflective of the actual customers in each segment. 

Mini-case study: Amazon masters iterative segmentation

One of the clearest examples of successful iterative customer segmentation can be seen in how Amazon tailors the shopping experience to specific and detailed customer segments. Behavioral data combined with a built-in feedback mechanism in the form of user product reviews gives Amazon the tools to segment, adapt and even anticipate customer needs. 

As a global company with more than 300 million active users, selling millions of products, Amazon uses geographic, demographic, behavioral and psychographic data to categorize customers. This means the brand can present products and services most likely to appeal to an individual. Parents of small children see offers for toddler toys and tools for organizing instead of ads for the trendiest club fashion, for example. 

Each time a customer looks at Amazon, clicks a link in an email or an ad, makes a purchase, leaves a review or otherwise interacts with the brand, information is added and the segmentation refined. This iterative and ongoing process is one of the reasons Amazon continues to be one of the most profitable companies to ever exist. 

Your content marketing strategy can be iterated, too. 

Messaging, visual hierarchy, calls-to-action and other elements of content can be tested and improved incrementally over time. Techniques such as A/B testing or cross-posting, among many others, offer methods for testing and improving content. 

Before creating the first video, writing a blog or social media post or recording a podcast episode, marketing teams need to consider the why behind the content. What is the ultimate goal? Increasing brand awareness? Encouraging more app downloads? Increased sales of a specific product? 

Once the answers to those questions are clear, content can be created and distributed more efficiently. Measuring the response is the critical next step because metrics direct iteration. Do people engage with the content through comments, reviews or click-thrus? What is the tone of the feedback? 

In short: Create, distribute, measure and monitor, then adjust and adapt. 

Changing the distribution channels, the CTAs, the messaging, the images or placement of the images or even a single word in a subject line are all examples of incremental changes. The whole process begins again by measuring the impact of those small changes and adjusting. 

Eventually, the content strategy may look completely different from the first draft, but the results should align more closely with the business objectives with each iteration. Often, using iterative testing in content marketing (or marketing in general) allows you to see trends ahead of the curve and make changes to take advantage of emerging market conditions.

CPG and QSR are two of the fastest-moving industries in the world. Consumers are fickle; tastes change, fads come and go, seasonal promotions and products are short-lived. All of that amounts to limited campaign lifespans. Brands must be responsive and implementing iterative testing makes responding a vital part of the process and the culture. 

Traditionally, it could take weeks for a brand to test a new package design, flavor profile or to determine if a content strategy was successful. Iterative testing means that marketing teams start getting information about what works and what doesn’t in hours and can implement (and test!) improvements faster. The entire research-plan-distribute-respond cycle is far shorter, giving brands in the fast-moving worlds of CPG and QSR an edge in the marketplace. 

What makes an iterative campaign successful? Some metrics to consider: 

• Speed of launch

• Number of failed concepts

• Improving engagement, clicks, views, reviews, etc. 

• Progress toward objectives in a given period of time

No matter which way you look at it, adapting faster benefits CPG and QSR brands. Iterative testing increases the speed of success. 

Learn how Zappi helps CPG and QSR brands create ads and products that win with consumers. 

Iterative testing originated and became popular in the world of software development. 

In that industry, the traditional way to release products was called the “waterfall method” and it took months or even years. A product was developed, then went through testing and changes within different siloed departments sequentially. 

The industry moved to a method called agile development that includes iterative testing. Teams are cross-functional and integrated and products are tested and incrementally improved. Products are released and updated and timelines are dramatically shortened. 

Although software development and CPG and QSR are very different industries, it’s easy to see how the principles of iterative testing can be applied in market research, customer segmentation and content marketing. 

Here are some tips to begin implementing this faster and more adaptive approach: 

• Begin with business goals, and keep them central to the entire process

• Include qualitative and quantitative analyses in the research

• Set up a feedback loop early, and monitor it constantly

• Decide which metrics to measure early, and track them through each iteration

• Refine each concept, campaign or outreach based on feedback

• Repeat the process again and again, making incremental changes with each iteration

Along with recognized best practices, iterative testing has some known pitfalls to be aware of and avoid, including: 

• Analysis paralysis - The idea is to research, launch, test, iterate, repeat. Don’t get stuck on the first step. 

• Getting stuck on one metric -Just like you need a balance of qualitative and quantitative data, you need to look at your metrics as a whole. Focusing too much on one can throw off perceptions. 

• Misreading signals too early - Although we’ve emphasized speed as an advantage of iterative testing, it’s possible to misread the earliest signals. Make sure each iteration has enough time for consumers to respond. 

• Failing to build a culture of innovation - If your team doesn’t understand the value of iterative testing and innovation, it’s unlikely to be successful. If it’s a new way of working, it’s going to take some effort to change the culture and mindset in your organization. 

Iteration requires a mindset shift across the department and organization and a willingness to adapt and change quickly. It’s not a tactic, but an entirely different approach to how teams introduce concepts and products. 

Making a cultural shift can take time, but the results are better alignment between what customers want and what brands offer, detailed customer segmentation and campaigns that meet business objectives — and who doesn’t want that? 

Building cross-functional teams where researching, testing and responding happen from the beginning to the end of the campaign lifecycle can take time and effort, but the results contribute to overall business growth.