To maintain SaaS growth, sometimes less R&D is actually more
By Yi Wang
Readers of this article are also reading:
Basing remote salary on geography? There may be a less contentious solution
Executive Summary
Among all industries, the market of Business Software and Services has the highest rate of product innovation (over 40% of revenue from new products), and generates approximately $43.7 billion in revenue in total. According to McKinsey & Co., healthy SaaS (software-as-a-service) companies are also expected to maintain a 40% revenue growth rate, partially fueled by Percent of Revenue From New Products (PRFNP). This lofty industry-wide target creates an imperative for individual SaaS companies to pursue high new product sales goals in order to remain competitive.
Research published in the Journal of Product Innovation Management suggests that when a firm manages the number of different types of external R&D collaborators it uses to an optimal range (1-7 collaborators, depending on if they are regional, national, or international partners), they maximize their PRFNP. When optimized, this tactic can introduce an additional 10-20% in PRFNP.
Leaders within the SaaS industry should pay close attention to these new findings. If they are struggling to meet growth and news sales targets, they should consider how their firm’s excess variety of R&D collaboration partners may be negatively impacting their PRFNP.
Impact: for SaaS, new product sales may be THE missing KPI
New product innovation is critical for business to stay ahead of the competition. The new products often have higher profit margins, target new markets, increase market share via differentiation, and may decrease costs. Innovation is key to a healthy business long-term. First, consumer “needs and wants” continuously change, and firms need to respond to the changes by introducing innovative products. Second, technology is upgrading at a fast pace, and existing products might be at the end of its product life cycle. Third, commercial environments change and companies might want to capitalize on those changes.
From the innovation data of the 2020 annual business survey published by the National Center for Science and Engineering Statistics, 11% of the estimated 4.8 million for-profit companies with at least one employee in the U.S. had some product innovation in 2017 to 2019. The SaaS industry contains the highest Percent of Revenue from New Products (PRFNP) with over 40%. From Precedence Research, the market of Business Software and Services in the U.S. has a market size of $109.2 billion in 2023, which means new product innovation alone generates revenue over $43.7 billion.
This new research comes at an optimal time for SaaS leaders. McKinsey & Co has identified PRFNP as the prominent managerial metric, of which R&D partnerships are shown to be critical. According to a 2022 McKinsey Global Survey, tech companies with a high PRFNP while maintaining core competencies across functions grow faster than their peers, and half of the companies’ revenues in five years is expected to come from innovation that do not exist now. Responses from the McKinsey survey further suggest that new businesses built by large organizations currently generate $5 trillion in revenues and could grow to $30 trillion in five years (“New-business building in 2022: Driving growth in volatile times”, McKinsey Digital, November 2022).
Additionally, PRFNP is critical to enterprise value. Tech company respondents expect nearly twice the value can be generated for an enterprise from every dollar of revenue from new businesses versus core business revenues (“New-business building in 2022: Driving growth in volatile times”, McKinsey Digital, November 2022).
While SaaS companies are also expected to maintain a 40% revenue growth rate, only a small share of SaaS companies can achieve it - according to McKinsey research, barely one-third of software companies achieve the average 40%. The goal of 40% revenue growth is especially important for a business to maintain as the firm matures. Of 100 SaaS companies in the United States with revenues above $100 million analyzed in May 2021, the median revenue growth rate was only 22% (“SaaS and the rule of 40: keys to the critical value creation metric”, McKinsey TM&T, August 2021). SaaS companies are therefore hard-pressed to meet revenue growth simply by scaling existing products without also a pronounced PRFNP.
Research: There is such a thing as “too many” R&D partnerships
Firms collaborate with different and geographically diverse R&D partners to enrich their knowledge pool, increase diversity in new product creation, and as a route to access more markets. Consider the Salesforce AppExchange, for example, which allows SaaS R&D collaboration driven by customer design. Some may only be small iterations, but in total can lead to new product lines entirely, all sourced via R&D partnerships across different industries and countries (ie markets).
However, research suggests that collaboration can present difficulties such as distance issues, capacity constraints, high transaction costs, and knowledge transfer limitations due to over-specialization of those partners and their needs. A recent paper published in the Journal of Product Innovation Management explored the distance and number of collaborators in R&D and how that affects firm product innovation.
Specifically, the paper wanted to know how product innovation was impacted by a firm’s “collaboration breadth.” Product innovation in this case is defined as the percentage of firm's total revenue from new products (PRFNP). Collaboration breadth was defined as the number of different types of collaborators with which they partner, i.e., enterprise groups, suppliers, customers, competitors, consultants and commercial labs, universities, or governments, on an ordinal scale between zero and seven. This was then further differentiated by the geographic proximity of those partnerships, i.e., same region, same nation, or international. So importantly, this did NOT distinguish between the number of national or international markets entered, nor the number of partners of any one type, only the variety of partner types by location was at issue.
The analysis was of three datasets and eight cross-sectional surveys collected on UK companies over 2004-2020, resulting in 25,813 observations on 14,784 firms spanning manufacturing, construction, professional services, wholesale/retail, information and tech industries. On average, these firms had 4.5% of PRFNP with most falling in a range of working with 1-2 external R&D partner categories. Notably, universities and government agencies were by far the least common partners.
The paper found, unsurprisingly, a significant benefit to PRFNP for having at least one R&D partner (so from zero to one): roughly 1% additional PRFNP for regional R&D, 4% for national, and 5% for international. But the regression model also showed the return was not the same for each increase in the number of partner type. For regional R&D partnerships - moving up the scale between one and seven increased PRFNP by only another 0.1% to 0.7%. However, there was far stronger yet curvilinear relationship for the number of national and international R&D partnerships. The relationship was an inverted U-shape, which the researchers posited is the disparity of the positive knowledge collaboration effect and negative cost effect. The optimal number of types of R&D partners was found to be three for national and international to maximize firm PRFNP – rising and topping out at an additional 1-4% benefit before dropping down again after three partnerships.
Thus, fully optimizing R&D relationships across all three geography types offers an additional 20% in PRFNP compared to having none. However, as it is likely firms will have at least some R&D partner across the three geography types, more realistically it is that extra 10% found in the 1-7 range that could lift a typical company from 30% of revenues from new products to 40% simply by adjusting their number of external R&D partners across all three geography types.
Reaction: Setting firm-specific R&D collaboration moderation
Practically, when a firm intends to increase their Percent of Revenue from New Products (PRFNP), finding many R&D collaboration types across different many geographies is tempting, as they open up new potential products and markets. But this can also get out of control and have a significant negative impact to PRFNP. Leaders targeting this metric would be recommended to:
Day One: Start by surveying all business units of their ongoing R&D partnerships. Then categorize all R&D collaborators into types and geographies, and perform an initial comparison to the optimal partnership mix as relevant.
Next: Test if the increase in PRFNP applies to your company by modulating collaborations with various types of R&D partners in each type of geography. While the research suggests firms should max out the number of regional collaborations to seven, and for national and international collaborations at or under three, it may be their own optimal number is not exactly the same – yet likely there is a moderating effect to discover.
Finally: As collaborators change incrementally over time, carefully track PRFNP as a KPI of this effort and react accordingly. While SaaS specific KPIs, such recuring revenue expansion from existing users or percent of revenue from new users, are focused on what is happening now, PRFNP is also important to be considering what will happen next.
The main focus of this paper is the influence of the number of collaboration types on PRFNP, but in reality other qualitative factors may also play an important role in changing PRFNP. For example, the degree of interaction with each partner, and the weighted influence from the seven types of partners, may vary results. Thus choosing which to add or cut may not be a one-for-one equivalent and SaaS firms should give extra focus on their core-competency partnerships.
____________________________
Yi Wang is a PhD candidate in Applied Physics at Northwestern University and the President of the Northwestern Advanced Degree Consulting Alliance. The research applications proposed in this article are solely the views of the author and do not necessarily reflect the views of the original academic journal article authors nor any individual member of our Editorial Board.