The State of Australian Software Product Management
Software product management plays a crucial role in the success of software companies. Product managers act as the voice of the customer, prioritise features and requirements, and facilitate communication and collaboration across teams. Effective product management fosters innovation by expanding market differentiation and exploring new technologies like generative artificial intelligence (AI). We’re seeing product teams like those at Alida and Canva embed generative AI technology to rapidly improve product capabilities. We’re also seeing product teams adopt generative technologies to drive operational improvements to software product delivery: Switchboard MD uses Amazon Q Developer to reduce the time to deploy new product features by 25% and Appian streamlined customer service for its product with an AI knowledge assistant enabling customers to pose natural language queries. At Amazon, we’re using generative AI to create synthetic test data to increase the accuracy of Amazon One.
While the opportunities to improve the operational efficiency of software product delivery are promising, It can be challenging to know where to start, and where the best opportunities for performance improvement and toil reduction might lie. With this in mind, we recently surveyed product management executives and practitioners at more than 100 Australian software companies, to get a "pulse check" on where product teams are investing their time, evolving approaches to product delivery, and the challenges that slow innovation and product differentiation.
Across all company sizes, we found that a majority of product management teams (52%) spend the most time in development (see chart 1). But one team in five (20%) actually spends more time in testing and validation. Time spent in testing and validation grows along with software company size. Almost one in three survey respondents (30%) say their product management teams spend the most time during the development lifecycle in the testing and validation stage. We think this reflects the increasing complexity of releasing products as software companies grow. As teams integrate multiple products into suites and build platforms of common services there is a tendency for product to become more monolithic, and when testing isn't highly automated product delivery velocity can slow. Investments in loosely coupled systems, automated testing, generative testing, and test avoidance can help.
This distribution of time allocation in our survey may indicate a need for PM teams to strike a better balance between development-focused activities and strategic, customer-centric activities like research, ideation, and go-to-market planning. That's especially the case when there are disruptive technologies like generative AI that change the way customers interact with our products (e.g. via Natural Language Query or Natural Language Generation). Dedicating more time to these areas will help ensure that products remain aligned with customer needs, ultimately driving higher product success rates through continued product differentiation.
Constrained resources are the biggest challenge overall for product management teams (54% -see chart 2): It's the top challenge for product teams at small (64%), medium (60%), and large (73%) companies. Accumulated technical debt is the second-highest challenge, with 36% of respondents identifying it as a major issue. And while managing dependencies amongst teams was the third-highest challenge overall (34%), it was most challenging issue for PM teams at the largest companies (44%). Given that getting executive buy in and making product decisions is a low challenges for survey respondents, we think this indicates that the challenge of dependency management is most likely related to highly coupled application code across product/platform sub-systems and the need for multiple teams to coordinate their integration activities. When coupled with percentage of product teams above that spend more time in Testing and Validation than any other product delivery phase it indicates to us that the management of multi-product code bases at large companies risks cutting into product innovation capacity.
Software companies use a variety of go-to-market models to reach customers. Early stage companies are often founder-led, but as they grow they embrace a sales-led or product-led model to drive product growth. In our survey, we see an even split between sales-led and product-led software companies (see chart 3).
Customer feedback, user data, and product data are the top three data sources product teams use for roadmap decisions, regardless of whether the company is product-led or sales-led (see chart 4). However, there are notable differences in the emphasis placed on certain data sources between the two approaches.
Product-led companies tend to rely more on user data, product data, market research, and performance metrics, suggesting a stronger focus on understanding user behaviour, product performance, and market trends. In contrast, sales-led companies place greater emphasis on sales data and competitive research, aligning with a sales-driven approach to product development. Organisations moving to Product-led Growth should consider what data points are essential for understanding user behaviour, product usage patterns, and customer journeys as well as grow skills required to perform data analysis and actionable insights that will informed good product decisions. We think generative AI technology can help here as well, whether it’s through personalized onboarding, question answering and self-service, or identifying product quality issues from trouble tickets and support requests.
If you’re interested in learning more about how product teams at software companies are innovating with generative AI both inside their products and as a way to drive operational improvements in product delivery processes join AWS, Xero and Canva on May 30 for an interactive virtual event AWS for Software Companies - Product & AI Exchange.
We’ll dive deep into practical ways to reduce time in software development through use of generative AI tools, giving software delivery teams and product practitioners more time to focus on customer obsession. We’ll also look at how generative AI use cases extend into product planning, go-to-market execution and sales activation to drive actionable insights and better product decisions.