Typically, there are different roles for ‘testing’ or research in positioning exercises: the search for unmet need against which to position (usually called ‘insights’), the evaluation of positioning options, and the improvement of a final option. This chapter will mostly focus on evaluation.

First, let’s establish this fundamental point: very few of us of us experience positioning in our day to day lives. Done well, it is only an internal construct that turns into things people experience. So, who are the people who can tell a good one from a bad one? Oddly, traditional testing takes this internal-only component and road tests it.

The way a product is presented or "framed" can greatly impact its perception. This is particularly true in the case of a new drug, where the framing can determine whether it is seen as a first-line treatment option or a last resort. The importance of a successful positioning is critical and should be approached actively, rather than passively.

As covered earlier in the book, one approach to positioning is the use of passive positioning. This is when data is simply presented and the audience is left to make their own conclusions about the product. However, this approach cannot then be effective in shaping the audience's perception of the product – it is a circular proposition, defining a product only as what is obvious and by what has come before - the audience’s existing perceptions. Instead, an active approach to framing is necessary.

Active framing involves making deliberate decisions about how to present the product in a way that influences the audience's understanding. This can involve highlighting specific features, comparing the product to others in the market, or presenting it in a specific context.

Market research is often used in the positioning of new drugs, but the traditional approach can lead to stereotyping and a lack of flexibility in the perception of the product. Specialists chosen for market research may have a preconceived notion of the product's place in the market, limiting the ability to shape the audience's understanding. They will typically have deep understanding of one field. What if that is the wrong field for your drug?

For example, describing a non-steroidal anti-inflammatory drug as a potential cure for Alzheimer's may initially confuse some physicians. The traditional perception of NSAIDs as ‘anti-inflammatory’ would limit their imagination and understanding of how it could affect a neurodegenerative condition. Perhaps their experience would tell them that other NSAIDs don’t. As soon as you frame the story, there is a limit to how far the physician will travel in her imagination.

On the other hand, presenting the ideas behind the product first (‘neuro-inflammation as a target, perhaps), rather than its category and its TPP, allows the audience to use their own imagination and projection, and provide a unique and compelling positioning.

The problem with passive positioning is that it's difficult to shift people's perceptions once they have been framed in a certain way. Traditional market research ignores that truth, and tends to rely upon passive positioning and quantitative differentiation as the core of its methodology, which can result in false positive views of a product's effectiveness. For example, imagine up-front in research showing a TPP that says a drug is 10% better than a current product. Presented with that ‘cold’ information, physicians may only then consider it for use in certain situations where incremental added benefit is sought. Is that 10% on a linear scale, does it provide an absolute benefit that couldn’t be achieved any other way, can physicians even tell the difference?

For example, not all improvements are perceived in a linear fashion in the real world. Kano's model is a great illustration of that. A popular tool for understanding customer needs and preferences, it can help organizations prioritize customer needs, evaluate the effectiveness of their product or service offerings, and develop strategies to better meet customer expectations.

Kano’s Model is a customer satisfaction theory that categorizes customer preferences into five types of attributes: Must-Have, Performance, Delighters, Indifferent, and Reverse. These attributes are used to analyze and prioritize customer requirements, and to guide product development.

• Must-Have attributes are the basic necessities that customers expect and take for granted, e.g. reliability of a product.

• Performance attributes are the aspects of a product that directly affect its quality, e.g. speed.

• Delighters are the attributes that exceed customer expectations and can provide a significant source of differentiation and customer satisfaction, e.g. style.

• Indifferent attributes have no effect on customer satisfaction, e.g. color of a product.

• Reverse attributes are the aspects of a product that, when present, negatively affect customer satisfaction, e.g. complexity.

At its core, Kano's model proposes that customers’ satisfaction with products or services is determined by three factors: basic needs, performance needs, and excitement needs. Basic needs are the foundational requirements of a product or service; they must be met in order to satisfy customers at the most basic level. Performance needs are those features that customers expect from the product or service; these features increase satisfaction as they are improved upon. Finally, excitement needs refer to features that delight customers; they provide an extra motivation to purchase and use a product or service. Now, imagine presenting a TPP up front in testing, and then asking physicians how they’d ‘position’ the drug. At its core, it is a collection of features, presented ‘cold’ - even a quick review of Kano’s model shows that a 10% change on the raw axes will make a different impact, depending on the feature being a must, a want or an exciter.

For example, you already know that a car that can do 35mpg is not as good as a car that can do 40mpg. Is the new option 14% better? Pharma would typically present it that way. What if this new car came with 5000 mile service intervals instead of the 10,000 of the older one? The way most pharmaceutical positioning testing works is by exploring the top-line mpg value to the respondent.

A lot of this research error is fed into the generation of positioning options. Unfortunately, the way that positioning is ‘tested’ can compound this mistake.

First of all, the idea of testing positioning is problematic. What exactly is being tested? As Deep Positioning is a strategic choice, it is important not to ask if the positioning being tested is a good ‘positioning’. Many research processes simply try to establish the likeability of a statement, or ask the physician to rewrite it to ‘sound’ better. As I said in the previous chapter, it is impossible to know if a positioning is good on first impressions. Testing in this fashion tends to prefer nice-sounding statements, especially ones that are vague enough to be widely interpreted.

Secondly, the choice of respondent is critical. Physicians know how things are today, and how things used to be. They will rarely know how the upcoming competitive landscape will look, especially if they have a narrow speciality (or are too broad – a typical challenge in finding research respondents). The likelihood that they are representative of the future audience is low. The likelihood that they will be able to understand their own motivations and decision process is also low, so a positively-presented statement will tend to do better. These two factors mean that testing positioning can lead to false positive views of its effectiveness.

So, why test at all? It might seem odd, but its main role is to give people who don’t understand positioning some confidence in the final positioning.

Typically, it is linked to the need for senior sign-off, addressed in the last post. Of course, that relies on two things: showing it is the ‘best performing’ option, and showing that it is likely to positively impact perception. This ‘false positive’ problem becomes an issue in this context. Nice-sounding ‘positionings’ do better in ‘testing’ and so tend to attract more easy sign-off, despite having no real world relevance.

Testing positioning options can be problematic in several ways, including:

1. False negatives: Testing may not detect a positioning problem even when it exists, leading to misdiagnosis or missed opportunities for improvement.

2. False positives: Testing may detect a positioning success that doesn't actually exist, leading to undue confidence. For example, an option being just the best of three poor options does not mean it is a good choice.

3. Limited specificity: Testing may not be able to distinguish between different types of positioning, making it difficult to determine the best course of action. Testing a bold, effective positioning that relies on an understanding of the disease that can only come from market shaping/ education against a ‘tagline plus’ positioning will inevitably favour the nice-sounding and easy one.

4. Observer bias: The results of positioning testing can be influenced by the person performing the testing, leading to inaccurate results. It is often unclear why a certain agency was chosen, but it is clear that the same outcome would not be achieved were different agencies to be applied - this means that the outcome is not necessarily objectively the best, but the one favoured by the agency chosen.

5. Inter-observer variability: Different people may interpret the results of positioning testing differently, leading to conflicting conclusions. It is rare not to have all observers used the words ‘I think…’ when detailing their conclusions, which suggests that objectivity is missing.

6. Unreliable results: Positioning testing can be subject to measurement error, leading to unreliable results and difficult-to-interpret data. It is impossible for it to be representative, so it can only be suggestive. One issue remains the averaging of the answers - the ‘average’ of 50 people hating an idea and 50 people loving it is not ‘indifference’.

7. Inconvenient or uncomfortable: Some positioning testing may be invasive or uncomfortable for the person being tested, which can limit the willingness of people to undergo testing. This assumes they were even the right kind of people to be in the testing. It also assumes that they can mentally project themselves into a future in which a whole lot has changed.

One way in which testing can be useful is in looking for objections, or unintended consequences. Words that seem perfectly innocuous in constructing a positioning can mean very different things to different people - it is useful to hear that in research before they’re used in the real world. For this to be useful, the research really has to listen, not be looking solely for improvements, or for a winner.

That kind of testing is exemplified in the process of improving prototypes, covered in the chapter on ideation and prototyping - you’re looking for the broken parts.

It is hard to find any good defence of testing for positioning. The test of a positioning is whether it makes a difference in the market. Few of the best positionings were ‘liked’ in testing, although they did score highly for memorability, differentiation and more.