質の悪いソフトウェアがもたらす隠れたコスト

AI is accelerating the software experience crisis

AI has made “coding” more accessible to the masses. But a truly great product is so much more than code. As development becomes faster than ever, the software experience crisis will worsen thanks to the rapid adoption of no-code development tools, citizen developers, and AI copilots.

The modern enterprise’s disjointed, siloed software landscape will also make it more difficult to use AI effectively. AI agents, like chatbots and digital assistants, only work if they can access all the data sources they need. If your essential data is siloed across dozens of software solutions, it’s nearly impossible to realize the full potential of AI.

Part 2: The five horsemen of the software apocalypse

Bad software experiences may not have a line on the balance sheet, but they cost companies billions yearly. Here are five key signs your company is dealing with a software experience problem:

Horseman #1: Low adoption and R&D waste

Pendo research found that as much as 80% of a software solution’s features are rarely or never used. And it gets worse: Nearly half of all business software goes unused, wasting an eye-watering $44 million every month. Ouch.

This financial waste is due to several factors:

• Companies can’t see how users navigate the software they’ve purchased.
• Users find tools hard to navigate, leading to frustration, decreased productivity, and shadow IT.
• Users don’t have the context or information to learn and use software to its full potential.
• Users don’t adopt core features or see the value of software, leading to disengagement and churn.

Whether in IT or product management, you invest in software to deliver a return on investment. But it’s clear that right now, organizations are overspending on tech and under-delivering on results.

Horseman #2: Security threats and compliance risk

A confusing software experience can put your company’s cybersecurity and brand reputation at risk. Security threats can spring up in many different ways:

• Poor software onboarding and training, which pushes employees to seek unvetted shadow IT solutions
• Confusing user interfaces (UI) can lead to misconfigured apps, ignored features, and unused security features
• Ignored, dormant apps that become an open door for hackers over time

The recent Change Healthcare ransomware breach is a prime example of this risk. Experts believe that hackers entered Change Healthcare’s software because a remote access system had multi-factor authentication (MFA) disabled. This means a Change Healthcare end-user likely chose to turn off that security feature for a more convenient login experience—but it cost the company $22 million.

Poor software experience can also lead to compliance errors and steep fines. The cost of GDPR non-compliance starts at €10 million, or 2% of a company’s annual revenue, but regaining consumer trust may prove even more expensive.

Horseman #3: Lower productivity

The productivity cost of cluttered tech stacks and poor software experiences cannot be overstated. Research shows that 69% of workers waste a full hour every day simply navigating between applications.

This single hour of daily productivity loss compounds to six weeks per employee annually—time directly attributable to inadequate experience management and the failure to strategically optimize cross-application workflows for efficiency.

To make matters worse, nearly one-third of employees say that poor software experience was a primary driver of their decision to leave a company. Beyond the immediate productivity drain, this attrition forces companies to spend more on recruiting, hiring, and onboarding replacement talent.

More troubling still, these new employees encounter the same problematic software environment that prompted their predecessors’ departures, perpetuating a costly cycle of diminished productivity and continued turnover.

Horseman #4: Innovation stagnation and opportunity costs

Every team makes technology investments with hopes of substantial returns: Product teams want a 20x ROI, IT departments want a 200% productivity increase, and revenue teams want to exceed their forecasted ARR.

Product leaders depend on insight into the user experience to inform their roadmap, prioritize resources, and retain customers. Without behavioral insights guiding your investments—like usage analytics, customer feedback, and session replay data—prioritizing your R&D investments is a guessing game.

Remember Snap’s Augmented Reality (AR) Enterprise Service unit? The team built this to serve their biggest customers, but enterprises were slow to adopt the service. Then, less than a year after launch, the entire division was shut down. If Snap laid the proper groundwork for product discovery, this might have been avoided. In times like these, let quantitative and qualitative data be your north star.

Horseman #5: Customer churn and revenue leakage

Your customers inevitably interact with your company through digital touchpoints—whether an app or website—turning every company into a software company. The quality of your software experiences directly impacts business metrics like customer satisfaction, churn, and lifetime value (LTV).

Never underestimate the impact of your digital experiences. It’s estimated that 70% of all online purchases are abandoned due to bad UX, leading to $1.4 trillion lost globally to preventable UX-related churn.

To prevent this, leaders need to know exactly how users engage with your software, how well your development roadmap fits their needs, and how you can set users up for success.

Business leaders also need to understand how employees interact with their tech stack. Poor software experiences can lead to employee burnout, but positive software experiences improve morale and reduce turnover.

Without visibility into quantitative and qualitative product data, you won’t know why customers are leaving—or be able to identify at-risk users before they churn.

Part 3: Why companies can’t ignore software experience management

The old software development lifecycle (SDLC) was never built for the speed and scope of AI. This once-in-a-generation opportunity to change how we build software comes with major improvements—and risks.

The modern software development lifecycle is slowing us down

Whether you’re building an app for customers or modifying software for employees, the old SDLC leaves companies blind to necessary improvements. Companies need to move to the modern SDLC: an iterative process that lets you analyze your software data and then act on it.

The best way to build great software experiences is to ground development in a deep understanding of your users. From discovery and validation to post-launch optimization, you can stay closer to your users and deliver better experiences with this nine-step process:

1. Plan: Define your project, business goals, and timeline.
2. Discover: Analyze quantitative and qualitative data to understand your users. Then, based on user feedback and data, validate the best solution.
3. Design: Translate these requirements into a development blueprint.
4. Implement: Create code based on design plans.
5. Test: Ensure the software you built meets performance and user requirements.
6. Launch: Release your software in a staged rollout.
7. Analyze: Monitor adoption, engagement, stickiness, and user sentiment.
8. Act: Fix vulnerabilities, resolve friction, and close the feedback loop with relevant users.
9. Improve: Continuously iterate and optimize your software.