How Do I Know If I Have a Good Idea?

Nowadays, there are more and more ways for Marketers to develop new consumer packaged goods ideas. Through the use of strategic brand studies, qualitative research, and co-creation methodologies, Marketers can generate a wide array of new ideas, often quite quickly and cost effectively. With a multitude of ideas to choose from, the real challenge is now deciding which ones should be fast-tracked, which should be fine-tuned, and which should be abandoned. In order to meet this challenge, Marketers need to assess the market potential for their ideas - in other words, the size of prize. It is not enough to know that a certain idea meets hurdles on survey-based measures to decide whether the idea should proceed into concept development. If the idea does not have enough market potential to meet business objectives, then either it needs to be improved, new ideas need to be generated, or the broader strategy needs to be re-evaluated.

Cracks in the System

The problem today is not a paucity of ideas. There are lots of ideas out there, but there aren't enough good ideas. Instead there are lots of mediocre ideas that clog up the system. What is needed is a better approach for evaluating ideas. At many companies, ideas are converted into concepts prior to identifying which ideas have the greatest potential for success. A process is needed to evaluate ideas before concepts are tested - a process which moves beyond idea ranking and incorporates objective benchmarks and market potential assessment.

The Right Start

Before embarking on the idea evaluation journey, it is important to differentiate between an idea and a concept. For practical purposes, an idea is a short statement that describes a product or a benefit. An idea does not include executional elements typically found in a concept such as a consumer insight, branding, positioning, emotional benefits, price, reasons to believe, taglines, varieties, or even visuals. An example illustrating the differences between an idea and a concept is shown in Figure 1.

Figure 1

So, why should you evaluate ideas prior to engaging in full-blown concept testing? The reality is that, in general, Marketers evaluate too many concepts making minor variations to the wording and execution in an effort to meet success criteria. The result is a waste of time, money and effort. It would be much more efficient to evaluate the ideas first, eliminating the weak ones and improving the strong ones, before moving on to concept building and testing, and then to field less concepts later on. In addition, we all know that adding executional elements to a concept can sometimes help it or hurt it; causing otherwise weak ideas to move forward because they are "carried" by a strong brand name, for example; or conversely, a strong idea may be killed prematurely because of a high price or poor choice of varieties. In short, we want to get to the core idea to see if it is compelling.

Increasingly, companies are recognizing the importance of idea evaluation but may not be employing a rigorous process. One qualitative approach that is sometimes used is to gather internal and external experts to pick the winners using a less than objective, non-empirical, "power-dotting" approach. For example, ideas may be posted to the walls of a conference room and everyone in the room is asked to stick dots on their favorite ideas. Of course, many participants look around to see where everyone else is placing their dots and then follow suit; this is often referred to as a "herding mentality." True independent evaluation is nearly impossible in these settings.

Another popular approach is to employ consumers in a co-creation process exercise in which participants are asked to create their own ideas and evaluate the ideas of others using a voting process. This approach is well-suited for idea generation because it employs large samples while leveraging consumers' creativity. However, it should only be used for a preliminary assessment of which ideas are best as the sample and questionnaire design are not sufficiently robust for determining which ideas should progress to the concept stage.

Beyond power-dotting and co-creation voting exercises, many companies are moving forward and conducting true quantitative idea screening in which large, representative samples of consumers are asked to rate ideas based on key measures. This is certainly great progress in improving the quality of innovation and making the process more efficient.

Unfortunately, most quantitative idea testing approaches are still rather primitive in that they contain several drawbacks, as described below.

Drawback #1: Use of Purchase Intent as a Key Metric

Many idea testing techniques use Purchase Intent as the key metric for evaluating ideas, not recognizing that Purchase Intent is not an appropriate measure at the idea stage. With Purchase Intent, unique ideas are often penalized due to lack of familiarity, while many ideas similar to current products are rewarded, thus fostering a "me-too" mentality. Moreover, it is difficult for consumers to employ a "purchase mindset" when evaluating ideas since they do not contain branding, pricing, or positioning.

Drawback #2: Lack of Benchmarks at the Idea Stage

By and large, solid benchmarks do not exist to determine which ideas are strong or weak. Most idea testing approaches are relatively new and database comparisons are not yet available. Relative comparisons amongst ideas tested are possible but one cannot assess whether all of the ideas are strong or all of the ideas are weak.

Drawback #3: Failure to Predict Market Potential (i.e.,"The Holy Grail")

Another drawback to most idea evaluation approaches is that they offer no systematic approach for determining size of prize, that is, in-market potential. Instead, guesses must be made and often those guesses are too high.

A Superior Approach for Quantitatively Evaluating Ideas

Best Measures for Idea Testing

Ipsos Marketing has developed a global idea screening tool called INNOScan that prioritizes ideas in terms of which ones are most likely to succeed as fully articulated concepts. INNOScan evaluates ideas on three key measures:

• Relevance...solves an unmet need
• Differentiation...does it in a unique way
• Believability...in a way that consumers know it works

Why should these three measures be used for idea testing? A best practice approach for idea evaluation is to adapt proven methods employed later in the innovation process for concept evaluation and forecasting. Ipsos Marketing has shown through extensive R&D and validation that Relevance and Differentiation are key drivers of new product success. Specifically, when a new product concept is perceived by consumers as both relevant and differentiated, it's almost sure to be a hit - with a 98% success rate based on our DESIGNOR database analysis. By understanding how to make your new products more relevant and more differentiated, you can substantially increase their potential. The same logic should be applied to idea testing. In fact, Relevance and Differentiation are especially well-suited for idea testing. Unlike Purchase Intent, Relevance measures the quality of the core idea by answering the question: does the idea meet consumers' needs? Differentiation lets you know if consumers perceive your idea to be different from other products or solutions currently available to them. And, Believability is an important diagnostic because it identifies which ideas will need to be supported with strong reasons to believe in order to convince consumers to try the product.

The results of an INNOScan study are provided in a dashboard format which includes a ranked Opportunity Score (overall composite score) and the three diagnostics (Relevance, Differentiation, and Believability) which drive the Opportunity Score (see Figure 2).

Quintile benchmarks are provided for the Opportunity Score and the diagnostics. A Hierarchical Bayes regression analysis is employed to provide more precision when the Opportunity Scores for the best ideas are very close.

Figure 2

In this example, there are two ideas with Opportunity Scores in the highest quintile of the INNOScan database. Both these ideas score in the top or second quintile on all diagnostics and are candidates to proceed to concept screening. The next three ideas score in the second or third quintiles overall. These ideas should be improved before progressing to concept screening. The remaining ideas are weaker and should not be advanced.

Benchmarking Idea Testing Results

In the past two years Ipsos has evaluated 6,000+ ideas and 10,000 concepts. Through R&D using both our idea and concept databases, Ipsos has uncovered a systematic relationship between the Relevance and Differentiation scores of ideas and concepts. So now we can provide more than a ranking of ideas; we can actually provide an objective assessment of the quality of the ideas themselves.

It should first be noted that ideas score lower than concepts because they do not contain the key positioning elements of concepts. In that sense, true idea screening is actually a tougher hurdle than concept screening. Given the strong relationship between scores on ideas and concepts, it becomes possible to leverage a much larger database of concepts for the benchmarking of the ideas. Now, the evaluation of a new idea based on 150+ consumers can be evaluated based on the learning of 16,000+ new product ideas and concepts in our databases. This type of evaluation provides a more accurate assessment than is provided by a small number of experts or consumers or by a quantitative tool without a large database and enables one to evaluate an idea against an objective standard of success rather than solely among the ideas tested. After all, we aren't just looking to identify the best of a number of bad ideas; we want to determine whether we are truly seeing some good ideas or not.

Forecasting the Market Potential as Early as the Idea Stage

Now onto the Holy Grail. Although the ability to leverage tens of thousands of other new products is a major advantage of the INNOScan approach, an even greater advantage is its ability to provide size of prize estimates - in other words, forecasts. The key is to leverage the benchmarking approach just described and then adapt a proven forecast model for concepts and products - such as the DESIGNOR174 model which has 600+ validations - to idea screening. With DESIGNOR, the critical information needed to develop a forecast for idea screening are some facts about the level of penetration, domination and fragmentation in the market, (1) a very simple or standard marketing plan, and scores on Relevance and Differentiation. With this information Ipsos can provide a rough estimate of trial or volume to see if the idea has enough potential to warrant further investment (see example in Figure 3).

Figure 3

In addition to the forecast for the idea, the deliverable includes a benchmark forecast for an average performing idea (Par Forecast). In this example, the Idea Forecast is higher than the Par Forecast because the idea performance is in the second quintile, which is above average. The Idea and Par Forecasts taken together provide a much better way to assess whether the idea will exceed, meet, or fail to meet sales objectives. Also, these forecasts provide guidance on whether sales objectives are reasonable or whether introducing any products in the market should be re-evaluated - essential learning for business strategy. Assuming the idea performs well enough versus objectives, then once the idea moves further along in the innovation process, the forecast can be updated to take into account positioning, branding, pricing, lineup, packaging, and copy with an updated marketing plan. The process is iterative and brings a strong financial perspective to innovation early and throughout the process. This approach greatly enhances the value of idea testing research and brings idea evaluation into the boardroom of the organization.

Conclusion

In order to gain a competitive edge, Marketers are increasingly seeking quantitative market research guidance at the earliest stages of their innovation processes. Frustrated by exhaustive concept screening and testing, the savviest Marketers are evaluating their innovations at the outset of new product development - the early idea stage. INNOScan allows ideas to be tested at a very early stage with a rigorous benchmarking approach. In addition, INNOScan provides the consumer inputs which can be employed to generate size of prize estimates. As a result, Marketers can deploy an efficient system for evaluating ideas that is better, faster, and cheaper than current qualitative or quantitative approaches.