A Governance Framework for the Idea-to-Launch Process

Development and Application of a Governance Framework for New Product Development

OVERVIEW: Many companies use some form of idea-to-launch process with staged gates for their new product development (NPD). This article reports the development of a governance framework for improving NPD decision making in complex product portfolios with short product life cycles. Refining governance Stage-Gate controls in six brand businesses in the footwear and apparel industry resulted in notable productivity improvement with management better able to find the "sweet spot" that balances creativity and control. A significant development is that cash profit growth is being planned from smaller product portfolios. The six brands reported an aggregate 49 percent increase in cash profit per product. Improving the level of governance controls in NPD can deliver improved portfolio cash margins and build greater confidence and conviction in product selection.

KEYWORDS: New product development, Portfolio management, Feedforward controls, Performance management

Refining controls in the Stage-Gate process can lead to significant improvements in project portfolio performance.

The global branded footwear and apparel industry, worth an estimated $300 billion annually, faces growing macroeco- nomic and industry-specific challenges. The key macroeco- nomic challenges are labor and material cost inflation from the predominantly Asian manufacturing base, a reduction in Asian manufacturing capacity as capital and labor move into different industries, and a lack of growth combined with retail price deflation in Western economies. Industry-specific challenges are the dynamic, volatile, and unpredictable nature of demand; short product life cycles; high levels of product variety and frequent product range changes; and complex, extended global supply chains requiring quick responsiveness to trends (Fernie and Sparks 1998; Bruce and Daly 2004). Against this complex backdrop, a significant and growing challenge for the industry is the increasing cost of excessive and underperforming innovation-innovation that does not enhance performance and hence is superfluous rather than functional (Bisbe and Otley 2004). This can happen when excessive new product development dilutes overall profit margins, design and development costs exceed the product margin, or large numbers of low-volume products signifi- cantly damage relationships with vendors.

To grow profits and avoid underperforming NPD, a firm managing large, complex product portfolios through a Stage- Gate NPD process needs to find the "sweet spot" that bal- ances creativity and control (Peters and Waterman 1982, 318; Cowen and Middaugh 1988; Simons 1994; Bisbe and Otley 2004). Finding this balance is crucial to driving portfolio productivity and improving the financial returns on the de- sign and development investment made in building the portfolio. This requires effective governance controls for portfolio management during the NPD process.

Governance in this context is defined as the manage- ment and control of the NPD control system. The effort to improve governance controls must recognize that different levels of quality and sophistication may be applied to evalu- ation and screening criteria, target setting, and alignment to strategy (Cooper and Edgett 2003). Consequently, the stan- dard of governance is determined by the quality of the NPD control system. The challenge is to identify how the quality of the NPD control system, and therefore governance, can be improved.

Applying a systems perspective to the Stage-Gate process highlights the importance of feedforward controls such as planning and forecasting when making go/kill/hold/recycle decisions (Cooper 1990). By combining the idea of different levels of feedforward-control sophistication with manage- ment's validation of targets, we have identified a new con- cept, feedforward anticipatory control (FAC), that has a crucial role in the process. FAC can help management achieve a better balance of control and creativity in the portfolio to drive productivity and profit growth.

In this study, we develop a framework to assess an orga- nization's level of sophistication in applying FAC and ex- plore the role of FAC as an important element in the control system. We used a longitudinal study of a multinational and multibranded footwear and apparel business to assess the effect of different levels of sophistication of governance controls on performance. Our findings show that raising the levels of FAC when managing complex NPD portfolios helps avoid wasteful product development and grows overall cash profits. The aggregate portfolio cash margins of the six brands in this study increased by almost 15 percent, and the cash profit from each product improved.

Stage-Gate Meetings and Feedforward Controls

In the branded footwear and apparel industry, the typical key stages in product-range building are seasonal strategy planning, concept development, product specification devel- opment, physical prototype development, final range review, and launch to market. Because of the seasonal nature of the market, the stages follow a cyclical pattern. When one range is launched, another one is already being developed to fol- low it.

A systems approach can help clarify the function of man- agement controls in the NPD process. A systems approach dic- tates that there are input and outputs and feedback and feedforward loops at each stage of the process (Wiener 1950, 1953; von Bertalanffy 1950) . Koontz and Bradspies (1972) studied feedback and feedforward control systems; adapting their model with the inputs, outputs, and control loops used in an NPD portfolio review meeting produces a control systems representation of the Stage-Gate meeting (Figure 1).

The inputs of the system are the data NPD managers bring into the review meeting, such as brand and product innovation strategy and a portfolio of products. The out- puts materialize when a product moves to the next gate or when the product achieves a market performance mile- stone. Other inputs to the system, often communicated from top management and acting as "boundary" controls (Simons 1994), can include target product margin per- centages or minimum cash margin targets per product. For example, management may declare that a product must achieve a minimum gross margin of 40 percent. These boundary controls can also include upper limits to the number of products to be developed.

A feedback loop is created when actual market perfor- mance data are captured and used to guide future NPD, creating new system inputs. Feedforward loops occur when information prompts action to avoid a deviation from planned performance or target outputs. Feedfor- ward controls are measures that look ahead to assess fu- ture performance, such as forecast cash profit for the portfolio. Feedforward control has been defined as "an- ticipatory control in which preventative action is taken before the difference between planned and actual perfor- mance occurs" (Ishikawa and Smith 1972, 166). In NPD, feedforward controls have been described in a number of ways: anticipatory control, strategic value planning, sce- nario planning, forecasting, goal setting, evaluation and screening criteria, feedforward controls, and feedforward metrics.

This presentation of Stage-Gate control combines two key ideas. First, in making go/no-go product decisions, managers can apply a feedforward control that anticipates the outcome of a decision, forecasting its future value. Second, when managers assess that future value against business targets, they can also question the validity of those targets (Argyris 1976, 1977; Senge and Fulmer 1993). This validity review may prompt a re-examination of the portfolio and the planned forecast. This combination of feedforward con- trols with target validation checks is FAC. FAC operates between the activities of evaluation and decision, at the point of decision, making it a vital element in governance control in Stage-Gate NPD.

In a two-phase study, we developed, refined, and validated a framework to describe how FAC works in the NPD process (see "The Study," above). The study began with a literature review to identify previous research on how managers use controls in NPD and how these controls enable or constrain NPD. This work revealed the importance of feedforward controls in enabling control without stifling creativity and identified five outline levels of sophistication in the use of FAC: no measurement, actuals reporting (Jørgensen and Messner 2009), forecasting and target setting (Makridakis 1986; Simons 1987; Bisbe and Otley 2004), scenario plan- ning (Miller and Friesen 1982; Karlsson and Åhlström 1997), and target review (Argyris 1976, 1977; Senge and Fulmer 1993). Beginning with these five levels, we refined our framework through observation of Stage-Gate meetings, interviews with NPD managers, and focus group discussions.

The FAC Framework

The refinement process produced a seven-level maturity framework (Figure 2). At level 0, brand businesses would not be doing any measuring or performance assessment of their range-building activity. (None of the brand group's businesses were in this position.) Level 1 is reached when feedback from the performance of current or prior ranges is considered in NPD decisions. Level 2 involves forecast- ing and target setting at the product category level. This helps management anticipate the performance outcomes for the new range on measures such as volume, sales rev- enue, and product margin. This is more than a high-level plan, as it forces managers to forecast at the product-category level.

Level 3 is a significant step change in control sophistica- tion. Forecasting is carried out at the product level, the lowest granular level. This product-level forecasting is car- ried out a number of times during the NPD cycle to help anticipate market outcomes. There is also a validation and "fit" check at the product level (Saunders et al. 2005) to assess whether the different strategic and seasonal de- mands required of the product range are being met by the products under review.

Level 4 represents another step change in control, with the addition of reporting and target setting for productivity ratios. For the participants in our study, the most crucial productivity metric was cash margin per product, although some participants used unit volume per product, which is simpler to forecast. We have labeled this central productiv- ity ratio the FAC metric; it plays a principal role in planning and has significant impact on the control of the NPD portfolio. It helps management cut through to the heart of decisions, balancing control with creativity and focusing attention on desired outcomes. This enables the team to filter out exces- sive innovation.

At Level 5, managers also apply scenario planning, a risk management tool that can help the management team visu- alize the future. Scenario planning is used to manage the level of uncertainty and assess whether the planned product range will deliver financial and strategic targets. Managers use this higher level of feedforward control to weigh alter- native courses of action (Miller and Friesen 1982; Karlsson and Åhlström 1997) and to optimize the product portfolio.

The highest levels of FAC sophistication exhibit the use of target validity checks, first at the product-category level (level 6) and second with the FAC metric (level 7). The combination of a target productivity metric with an anticipated volume or cash profit outcome determines the optimum size of the port- folio. For example, a target portfolio cash margin of $10 mil- lion with a target average cash margin per product of $50,000 suggests an optimum portfo- lio size of 200 products ($10m + $50,000 per product = 200 optimal portfolio size). This type of calculation, based on a target productivity metric, guides decision making and further reinforces the crucial role that the FAC metric can play in governance control.

Each higher level of FAC sophistication is built on the application and learning from the lower levels. This means that attempts to operate at higher levels are unlikely to be effective unless lower level applications have been consolidated. As brand man- agement teams increase their sophistication in the use of FAC, they increase the cross-checking between short-term tar- gets and longer-term strategy, through simple validation at level 3, uncertainty managed with scenario planning at level 5, and target validation checks at levels 6 and 7. We therefore believe that a higher FAC sophistication level will create a bet- ter balancing of short- and long-term innovation needs.

The key purpose of this research was to understand how changing the levels of governance controls, and particularly FAC, affects performance. The branded footwear and apparel industry is ideal for this type of study given its relatively fast "clock speed." The key NPD stages will typically be completed in 26 to 52 weeks, with feedback from market performance within months of the product launch.

The Intervention

When we compare the hard numbers from the six brands where we intervened against the three control brands where there was no intervention, the difference is apparent. Brands where FAC levels were improved reduced the size of their product portfolios and made more money compared with brands in the control group (Table 1). All the brands with which we intervened are experiencing step-change improve- ments in portfolio productivity metrics. Cash profit and planned cash profit growth is being achieved at much smaller portfolio sizes. Across the six brands, the key season-total product-range size has reduced from 2,884 products to 2,210 products, a reduction of 23 percent, with cash profit up by 15 percent and cash margin per product up by 49 percent. Brand Six, the lowest revenue brand, at pre-intervention had the lowest productivity by a significant margin. Post-intervention, Brand Six has reduced the portfolio size by 65 percent and increased portfolio cash margin by 32 percent, resulting in a 275 percent improvement in cash profit per product. Across the brands, the number of underperforming styles is declin- ing and the product ranges are more focused. As a product manager told us, "We are freeing up capacity to do a much better job on fewer styles and improve margin." Another product manager reported finding the process "liberating be- cause instead of firing shots all over the place it's now one bullet that kills it dead." New rules of thumb are also being used in review meetings to manage portfolio size, such as "fewer, bigger, better," "one in, one out," and "do less and love it more."

By comparison, across the three control brands, the total product range size increased from 956 to 1069 products, an increase of 12 percent, with portfolio cash margin up by 8 percent and cash profit per product down by 4 percent.

In the brands where we intervened, managers took a number of actions to increase FAC levels. They acted to col- lect new performance information and enforce a closer focus on range-build performance and set new targets for product category range sizes and product productivity. These actions have produced other changes in the NPD process, including changes in the levels of new and carryover products in the portfolio, the development of analytical tools to assess prod- uct viability, and the move to more cross-functional Stage- Gate reviews.

Product management teams are now trying to understand better how many totally new products are required to meet targets, versus simpler options such as color or fabric changes or carryovers of prior-season products. Brand Sports shared with the other intervention brands a forecast margin- mapping approach that has resulted in changes to review meeting agendas. Another noticeable development across the intervention brands is a new aspiration among NPD managers for more collaborative product and category fore- casting, with greater cross-functional input starting earlier in the process. The more regular cross-functional reviews of forecasts seem to be driving greater questioning of the viabil- ity of product projects. Overall, managers involved in the NPD process clearly recognize the performance benefits of increasing the sophistication of governance controls.

Case Study-Brand Sports

Brand Sports sells sports apparel and equipment in over 100 countries. At over 60 years old, it is one of the oldest brands in the market, and it is well known for its innovative prod- ucts. Its product portfolio spans a broad range. At one end of the spectrum, its technical products are used by world- class athletes; at the other end, the brand's fashion and lei- sure products have very broad consumer appeal. The diversity of the brand's product lines is reflected in a portfo- lio made up predominantly of incremental innovation with some radical innovation (Figure 3); this composition is fairly typical of the other brands in the study, as well.

Before the intervention, Brand Sports typically designed and developed 850 products for launch into the market for its key season. At our initial review, Brand Sports was oper- ating at FAC level 2: managers captured and reviewed sales and margin performance at the product level for the prior seasons. They also generated revenue and unit volume forecasts, built up from the lowest product level. Together, these analyses established aggregate pictures for each product category. However, these forecasts were made only once, at the start of the NPD cycle, and only by product management.

The first key intervention with Brand Sports was the pre- sentation of the FAC framework. This was followed a few months later with a new analytical tool, developed collabora- tively by researchers and management, forecast margin mapping by product category (Figure 4). This tool helps man- agement map the portfolio of products from a margin perspec- tive, segmenting the range into more manageable cash profit groupings. When managers in Stage-Gate meetings review the ma r gin map, they consider key questions such as:

* Will this product enhance or reduce the overall target cash margin per product?

* How can we drive more volume from the "margin--rich" product?

* How can we improve the margin of the "volume-driver" product?

In answering these questions, the management team uses the FAC metric as a product margin threshold that needs to be exceeded by the products under review. Different portfo- lio scenarios are considered, and there is a formal validation check on the overall product category targets and FAC met- ric. Six months after the first intervention action with Brand Sports, simple bottom-up forecasts were reviewed cross- functionally three separate times from idea to launch, rather than once, which had been the norm.

A year after the intervention, by fall 2012, Brand Sports had reached FAC level 7. In the spring/summer 2013 season, it launched 658 products, a 22 percent reduction from previous levels. At the same time, the cash margin for the range has increased by 16 percent. The brand FAC metric of cash margin per product has increased by 49 per- cent. Cash margin per product targets have now been set for all product categories. The management team is man- aging a smaller portfolio of products that are performing better, both in design and development quality and in cash margin productivity.

Conclusion

In businesses with large product portfolios, managing prod- uct portfolio performance, using Stage-Gate NPD processes, is a crucial part of a key value-creating activity. Improving governance controls in the management of Stage-Gate product- review meetings can enhance NPD performance. FAC plays a crucial role in categorizing different levels of governance con- trol and the FAC metric to guide decision challenges in control versus creativity. In our study, applying more sophisticated FAC tools resulted in improved cash profit performance with a tighter product portfolio. The improvement in control is per- ceived to be building greater confidence and conviction with product selection. This creates product portfolios that better balance creativity with efficiency and the needs of short-term and long-term innovation. The next step in this exploration will be to capture the softer perception measures of perfor- mance change. This will enable a more comprehensive map- ping and understanding of the effect of changes in FAC sophistication levels on portfolio performance.

However, this study was conducted in the context of a large player in the global branded apparel and footwear industry, where the size and shape of the portfolio is important to the customer, and with predominantly incremental innovation. Industries with different product portfolio requirements may well have different FAC requirements. The study is also more appropriate to established suppliers; companies starting out of- ten have different problems in building their portfolios.

That said, there have been few studies of the practical application of governance controls in the idea-to-launch process. Longitudinal studies following the intervention and impact across multiple brand businesses is an appropri- ate way of developing our understanding. Other approaches can capture snapshots of current practice but following de- velopments through time can tease out the key elements providing insights into how performance can be improved.

The Study

To investigate the effect of FAC on NPD performance, we carried out a two-phase research project, working with eight brands in the sports, fashion, and outdoor footwear and ap- parel categories. The participating brands are subsidiaries of a multinational brand group that operate as self-contained busi- ness units. Together, they represent aggregate annual sales of $1.2 billion and employ just over 700 people. The brand group has been in existence for over 60 years; it is well known for the high quality of its sports and fashion products, its ability to track and sometimes lead fashion trends, and specific technical products that are used by world-class athletes.

Phase 1: Developing the FAC Framework

Three different methodologies were used to gather data and refine the FAC framework developed from an initial litera- ture review: observation of product-range review Stage-Gate meetings, semistructured interviews, and focus groups. Five brand review meetings involving 56 participants were ob- served for a total observation time of 20 hours. These obser- vations were supplemented by 18 semistructured interviews, including two interviews with small groups, for a total of 26 interviewees. The interviewees were all participants in Stage- Gate meetings and represented the different management functions involved. Three focus groups were conducted with a total of 37 participants, all managers involved in product review meetings. Focus groups were asked to discuss two questions: In the NPD process, what works well? And what doesn't? The FAC governance framework was refined as a result of these observations and interviews.

As a final step, to validate the framework, the findings were presented to 15 managers directly involved in Stage-Gate review meetings, all of whom reported finding the framework useful and easy to apply. Managers recognized that progression up the lev- els of governance requires the consolidation of lower levels; their comments suggested that the framework could be used as a roadmap for future development:

* Managing Director: "It's simple and clear. The beauty of having something like this is considering the perform- ance metrics we are striving for and all the time looking to improve. And we're trying to do all that at level 1. At that level, we're just not going to get that much higher performance."

* Merchandising Manager: "The timing is perfect for this. It's very good. I like the 'sophistication' terminology."

* Category Manager: "These are the stages we are going through. We are going up and down these levels. We have some foundations in place. It's a really good gauge for 'hon- estly, where are we?' This is a really good roadmap. We need robustness in the levels below before really getting into scenario planning.

* Product Manager: "This is what I feel deep inside. . . . At the end of the day, what you are giving me is the tools. It's a struc- ture, process, and next-step sense check. It's not 'finger in the air.' We can do more up front, more of a strategic business plan than a short-term vision, and doing the targets makes sense."

* Supply Chain Team Leader: "It's about doing less and get- ting more. I can instantly see the framework and levels and where we're at and where we're going."

Many of the managers made an almost immediate decision to adopt the ideas captured by the framework; several reported being motivated to reach higher levels of FAC sophistication. It is clear from managers' responses that those involved find the governance framework passes the practical benefit tests of usability and usefulness (Platts 1993).

Phase 2: Measuring the Effect of FAC Implementation

The second phase of the study involved intervening in six of the original eight participating brands. (During the period of study, one brand went through significant organizational restructur- ing and another was sold off; these two were eliminated from the study.) Having identified the existing FAC level for each brand, we carried out interventions to improve the level and use of FAC. The initial intervention involved presenting the FAC governance framework to the brands' NPD management teams. Each of the brand teams then received assistance in developing new feedforward metrics and analytics, and sup- port to implement improvements in FAC sophistication. The two lowest-revenue brands were supported with facilitation in product review meetings. To date, there have been 39 sepa- rate interventions involving 33 people across the six brands. We collected data on the change in FAC levels in each brand, conducted interviews with participating managers, and gath- ered financial metrics documenting the actual performance of each brand for the year following the study. Additionally, we collected comparative data on three brands in the same brand group for which no intervention was carried out.

Each higher level of FAC sophistication is built on the application and learning from the lower levels.

Applying more sophisticated FAC tools resulted in improved cash profit performance with a tighter product portfolio.

References

Argyris, C. 1976. Single-loop and double-loop models in re- search on decision making. Administrative Science Quarterly 21(3): 363--375.

Argyris, C. 1977. Double-loop learning in organizations. Har- vard Business Review 55(5): 115--125.

Bisbe, J., and Otley, D. 2004. The effects of the interactive use of management control systems on product innovation. Ac- counting, Organizations and Society 29(8): 709--737.

Bruce, M., and Daly, L. 2004. Lean or agile: A solution for sup- ply chain management in the textiles and clothing industry? International Journal of Operations and Production Management 24(2): 151--170.

Cooper, R. G. 1990. Stage-gate systems: A new tool for manag- ing new products. Business Horizons 33(3): 44--54.

Cooper, R. G., and Edgett, S. J. 2003. Overcoming the crunch in resources for new product development. Research-Technology Management 46(3): 48--58.

Cowen, S. S., and Middaugh, J. K. II. 1988. Designing an effec- tive financial planning and control system. Long Range Plan- ning 21(6): 83--92.

Fernie, J., and Sparks, L., eds. 1998. Logistics and Retail Manage- ment: Insights Into Current Practice and Trends From Leading Experts. London: Kogan Page Ltd.

Ishikawa, A., and Smith, C. H. 1972. A feedforward control system for organizational planning and control. Abacus 8(2): 163--180.

Jørgensen, B., and Messner, M. 2009. Management control in new product development: The dynamics of managing flex- ibility and efficiency. Journal of Management Accounting Research 21(1): 99--124.

Karlsson, C., and Åhlström, P. 1997. Perspective: Changing product development strategy-A managerial challenge. Journal of Product Innovation Management 14(6): 473--484.

Koontz, H., and Bradspies, R. W. 1972. Managing through feedforward control. Business Horizons 15(3): 25--36.

Makridakis, S. 1986. The art and science of forecasting. Interna- tional Journal of Forecasting 2(1): 15--39.

Miller, D., and Friesen, P. H. 1982. Innovation in conservative and entrepreneurial firms: Two models of strategic momen- tum. Strategic Management Journal 3(1): 1--25.

Peters, T., and Waterman, R. H. Jr. 1982. In Search of Excellence: Lessons from America's Best-Run Companies. New York: Harper and Row.

Platts, K. W. 1993. A process approach to researching manufac- turing strategy. International Journal of Operations and Produc- tion Management 13(8): 4--17.

Saunders, J., Wong, V., Stagg, C., and Fontan, M. M. S. 2005. How screening criteria change during brand develop- ment. The Journal of Product and Brand Management 14(4/5): 239--249.

Senge, P. M., and Fulmer, R. M. 1993. Simulations, systems thinking and anticipatory learning. Journal of Management Development 12(6): 21--33.

Simons, R. 1994. How new top managers use control systems as levers of strategic renewal. Strategic Management Journal 15(3): 169--189.

Simons, R. 1987. Accounting control systems and business strategy: An empirical analysis. Accounting, Organizations and Society 12(4): 357--374.

Von Bertalanffy, L. 1950. An outline of general systems the- ory. The British Journal for the Philosophy of Science 1(2): 134--165.

Wiener, N. 1950. The Human Use of Human Beings: Cybernetics and Society. London: Eyre and Spottiswoode.

Wiener, N. 1953. What is cybernetics? In Readings in the Philoso- phy of Science: Introduction to the Foundations and Cultural As- pects of Science, Ed. Philip P. Wiener, 100--110. New York: Charles Scribner's & Sons.

Mark Baker is a doctoral researcher in the Centre for Business Performance at Cranfield School of Management. His research is in management control systems and new product development portfolio management. He has a materials science and engineering degree from London University and is a Fellow of the Chartered Institute of Management Accountants. He worked for Unilever for 12 years in a variety of roles, including manufacturing, supply chain, and finance. Mark has also worked for 15 years in the footwear and apparel industry, again, in a variety of senior roles, including finance, supply chain, project management, and risk [email protected]

Mike Bourne is professor of business performance at Cranfield School of Management. His research, teaching, and consultancy are in the field of performance measurement and management and its application in control, decision making, and strategy execution. Before his academic career, Mike spent 15 years in business. He held a number of positions, with roles in production management, strategy and acquisitions, IT, HR, and commercial and general management. Mike is a chartered engineer and a chartered management accountant. [email protected]

DOI: 10.5437/08956308X5701105