Designing the Best Strategy for Your Next Global Product Rollout

Picture this: After investing significant resources into designing, testing, and building out a new product, you’ve finally cleared all major development hurdles and you’re ready to launch. Will you roll out the new product slowly to select customers and see how they respond before expanding further? Or will you offer it to everyone all at once? It’s a critical decision, especially in industries with high distribution costs and tough competition.

If you go all-in on a global launch, you risk substantial losses if the product doesn’t end up being as successful as you had hoped. Imagine going through time-consuming negotiations with multiple sales channels, cranking up supply levels, customizing packaging for each consumer segment, and paying upfront for countless other setup and distribution services, only to learn that these sunk costs will never be recouped.

On the other hand, if you launch slowly and selectively to guard against the costs of failure, you may be preempted by industry rivals. Testing prospective markets is important, but if you spend months studying customers’ responses, slowly dipping a toe into one small market at a time, your competitors can quickly grab market share. Even if a handful of customers love your product, giving your rivals the chance to flood the market with their own offerings can cost you your first-mover advantage.

Clearly, there are pros and cons to each approach — and it’s often far from obvious which will be optimal in a given context. For example, when launching their hybrid cars, Toyota stuck to the Japanese market at first, reserving the option to amend or scrap the Prius before launching abroad, while Honda launched the Insight simultaneously in the U.S. and Japan. Honda successfully beat Toyota to the American market, but its wider distribution meant failure would be more costly. Unfortunately for Honda, the Insight indeed turned out to be unpopular with American drivers, leaving room for the Prius later to eventually capture a much greater segment of the global hybrid market. Did Honda make the right decision and just get unlucky? Or should the company have gone more slowly, as Toyota did, and risk being overtaken by a competitor?

Experiment First or Go All-In?

Our recent research explores how firms can navigate this tension between launching all at once and step by step. We documented the rollouts of more than 300 new mobile phone innovations through a comprehensive analysis of thousands of models from more than two dozen manufacturers. This dataset was particularly informative because mobile phones are a massive, global industry, and yet their distribution must be managed on a country-by-country basis, making the all-in approach particularly costly.

We consistently found that being the pioneer of a given innovation in one market did not guarantee that a firm would be the first in other markets: In our sample, pioneers were overtaken in two-thirds of all subsequent markets. We also found that failure was far from uncommon: Eighty percent of the innovations in our study failed, suggesting some caution is called for when rolling out an untested product feature.

So, how should firms determine which approach will be the best fit for their launch? In our study, we saw that global pioneers of mobile phone innovations generally launched more selectively than followers did. This makes sense, since with limited precedent, there is more to be gained by careful testing, whereas when a market has already been validated by a competitor, there is less need for experimentation. For example, Sanyo lagged behind Kyocera in the introduction of the first-ever camera phone, and so Sanyo chose to launch more widely (including, crucially, in the U.S. market). In contrast, as a camera-phone pioneer, Kyocera was initially less worried about competition and more worried about failure risk, and so the company launched its camera phone more cautiously, in just the Japanese market.

Especially for technologies that advance quickly (which is true of many innovations in the mobile phone space), we found that there was a limited window in which money could be made from a given innovation. As such, as more and more firms began to offer the camera feature, companies became less cautious and more aggressive in how they approached their launches.

To be sure, not all the companies in our study successfully followed this logic. Especially for those who had recently experienced a major success or failure, it could be difficult to avoid being swayed by past experience: Firms were quicker to launch if they had been overtaken by a rival in their recent past, and slower if they had just been burned by a big flop. For instance, after competitors preempted Kyocera’s camera-phone innovation in international markets, the firm took a somewhat less cautious tack to rolling out its next innovation. On average, we found that firms that had recently missed rolling out a hit feature targeted an average of 137 million more consumers in their next innovation rollout than they had in their prior ones. Conversely, we found that just one big flop in the last year led mobile phone makers to target their initial rollouts at smaller audiences, with an average of 220 million fewer consumers. Making adjustments like these is an understandable response to a disappointing recent experience, but it often isn’t the best move for the firm. When deciding on a rollout strategy, lessons of the past are not relevant as the characteristics of the innovation and market at hand.

Good Testbeds Are Uncomfortable

Of course, determining your rollout approach is just the first step. The next is implementation. If you go the all-in route, the main challenge will be coordinating a complex global launch. But if you aim to test the waters first, you’ll have to determine where to start.

It can be tempting to launch in a market you’re familiar with, or where you already have significant market share. This maximizes your chances of immediate success — but it doesn’t do much when it comes to helping you learn more about your product’s broader market potential. Safe markets are uninformative testbeds. Your most loyal customers’ response to your newest product doesn’t tell you anything about its commercial viability in more-challenging environments.

The ideal testbed is one where customers can choose among competing offerings. That’s why the savvier experimenters in our dataset sought out challenging markets, where they could generate more-informative signals of global performance — even if that meant lower returns in the near term. Specifically, there were two types of signals that firms used to better understand a product’s potential:

• Robust indicators of success, which can only be gleaned from markets where it’s tough to succeed, and which help firms avoid erroneous rollout extensions
• Robust indicators of failure, which can only be gleaned from markets where it’s tough to fail, and which help firms avoid erroneous rollout terminations

Robust indicators of success come from markets where would it be difficult to succeed: Places where competitors already sell a version of your latest innovation. Although launching in such a harsh environment might not seem to make a lot of immediate financial sense, the learnings that such testbeds can provide may actually help you make the most of your innovation resources in the long run. For example, if Motorola wanted to be confident about the market potential of its camera phones, it would be best for the firm to launch in Japan or Korea, where it would have to outshine local powerhouses such as Kyocera and Samsung to succeed. If Motorola’s new camera phone sold well in those markets, it would likely succeed anywhere.

Conversely, robust indicators of failure come from markets where would it be difficult to fail. For example, a skincare company might pick Paris or Tokyo to launch a new product, since customers in these cities are generally particularly enthusiastic about trying out the latest new beauty products. When P&G piloted its SK-II range of makeup products in Tokyo, it could be sure that if it flopped, it wouldn’t be because local consumers were unaccustomed to multi-step skincare routines — if SK-II failed there, it would eventually fail elsewhere too, and so P&G would be able to terminate the product line with confidence.

Gathering both of these types of signals helped the firms in our dataset be more confident about their launch strategies, whether they decided to go all-in on a certain market or call it quits.

***

Of course, there are no one-size-fits-all solutions to planning market entry. In a prior study, we documented how earlier entrants to a given market generally benefit from trying out many different innovations, while later entrants are better off focusing on a narrower range of tried-and-true products or features. Our latest research builds on this finding by exploring the impact of different kinds of rollout strategies: An early mover interested in trying a wider range of innovations should launch sequentially, beginning with challenging test markets, in order to glean more information around consumer preferences and demand in different markets before expanding. In contrast, once a late mover has carefully determined the type of innovation that the market seems to want, it will likely benefit most from a simultaneous worldwide rollout. No one can escape market uncertainty — but choosing a rollout strategy that’s well-suited to the specific characteristics of your firm, innovation, and market can improve your odds.