Why Do IoT Projects Still Fail?

Posted by Andrei Klubnikin

The third edition of the Microsoft IoT Signals survey unveiled that 90% of companies around the globe have already adopted the Internet of Things solutions. 66% of the respondents, meanwhile, shared plans to incorporate IoT in more applications and workflows in the coming years.

However, the same research highlighted several persistent challenges in IoT solution development. 

For instance, 30% of organizations were still in the process of implementing connected systems, citing technical problems as a key barrier to wider IoT adoption.

Other companies (29%) have expressed concerns about the security challenges that arise from IoT implementation. Additionally, 26% of businesses reported a shortage of staff, while 24% lacked the necessary expertise to carry out IoT projects.

Things don't look too bright on the business side either.

Specifically, many executives do not clearly understand the Internet of Things’ value in the early implementation stages. This prompts technical teams to invest more time and resources in proof-of-concept (POC) projects and pause them to re-evaluate IoT's impact on a company's operations and business model.

Overall, up to 35% of all IoT projects are brought to a halt during the PoC phase due to high scalability costs (32%), lack of necessary technologies (26%), and unclear business value and ROI perspectives (25%).

Coupled with the massive IT specialist layoffs and stagnant economies, these insights prompted experts to lower the 2023 predictions for the global IoT market growth from 24% to just 18.5%.

Does this mean IoT is doomed?

I wouldn't be so pessimistic.

Any IoT project can be successful as long as you carefully plan and avoid some common implementation problems.

Why IoT Projects Fail: Top 5 Reasons & Tips to Avoid the Unfortunate Scenario

Technology-Related Challenges Manifesting Themselves Late in the Project

Connected solutions vary in size, scope, and complexity.

An inventory management system that uses Bluetooth tags is considered an IoT system. Similarly, a remote patient monitoring solution that includes custom wearable devices for tracking medication intake, physical activity, and falls is also classified as an IoT solution.

However, the functional and non-functional requirements for such systems would be sonically different.

Without collecting and analyzing these requirements before any development activities, it is impossible to select the appropriate technology stack for the software component of an IoT product and make important considerations regarding hardware design.

Unfortunately, many companies join the IoT trend without fully understanding how their connected systems will function behind the scenes. This leads to wrong technology choices and costly workarounds late in the development process, causing the project to go awry.

Solution

Partner with skilled business analysts (BAs) who have hands-on experience implementing IoT in your industry.

Next, start your project by conducting a discovery phase. This phase involves the collaboration of BAs, software architects, and hardware engineers. Its goal is to determine how the connected system will operate in various scenarios and to document these requirements. Additionally, a blueprint for the solution architecture will be created during this phase.

Equipped with this knowledge, you'll have no difficulty selecting suitable IoT communication protocols, setting up an efficient cloud infrastructure, and developing crystal-clear guidelines for the hardware and embedded software engineering teams.

The results? You won't find yourself in a situation where data exchange protocols drain your oil rig sensors' batteries too quickly or fail to relay sensor readings to the central hub.

Poor Communication Between IoT Project Teams

The Internet of Things solutions incorporate multiple technology components, from connected devices to cloud services and applications visualizing data in real time.

Few companies have the right skills, expertise, and specialists to design and implement IoT systems by means of an in-house team.

Outsourcing the development of IoT solution components emerges as a viable solution to this problem.

But there's a catch.

Halfway through the project, you may realize your embedded team cannot devise connectivity requirements for the mobile application developers, while the project manager (PM) on your side leaves for another company.

The consequences of team misalignment can be drastic.

You go over budget, struggle to meet deadlines, and end up launching a product that is either too complicated for target users or misses essential features to take off.

Solution

Your IoT endeavor should be guided by an experienced PM who's cut their teeth on distributed team management and excels in Agile.

The reason for choosing Agile as the preferred approach to connected solution implementation is quite obvious: in innovative projects, requirements and priorities often shift. This calls for certain flexibility, which can't be achieved with the Waterfall methodology.

To effectively manage multiple tech teams on your project, it is important to choose reliable project management software and establish clear communication processes that all parties involved should adhere to. Such processes may include regular team syncs, sprint retrospectives, and clear task prioritization.

Also, make sure that your company's stakeholders participate in important discussions, such as feature set prioritization and budget approvals. 

IoT Solution Scope Creep

When developing IoT products, particularly for the consumer electronics segment, you might be tempted to overstuff your product with fancy features.

The truth is that you may not have the necessary funding and expertise to build a smart home hub with voice control capabilities that is also compatible with Google and Apple HomeKit products and processes data locally.

Commonly referred to as scope or feature creep, the phenomenon causes up to 60% of all IT projects to fail, either partially or completely.

Solution

To accelerate the implementation of IoT, it is advisable for your company to embrace the "start small, think big" approach. This entails gradually building, testing, and deploying the core infrastructure of your connected system, and then adding more advanced features as your product expands.

This way, you'll be able to deploy the solution faster, start gathering user feedback straight away, and leverage the insights to continue improving your product. 

Opting for a minimum viable product (MVP) version of your IoT system, which includes only the essential features to meet initial demand and solve real-world problems, can help you gain approval from management for further development activities.

Overlooking scalability and hidden infrastructure costs

As we mentioned earlier, 35% of IoT proof-of-concept projects never make it to production due to high scalability costs.

What does this mean in practice?

Suppose you're building an intelligent controller that opens and closes curtains at predefined intervals. The system would send status data to the cloud at preset intervals, and every server call costs money.

The amount of data traversing the network (as well as the number of server calls) will increase proportionally with the number of system users. If you fail to configure efficient data caching mechanisms and reduce the number of server calls, your cloud infrastructure will become too costly to scale and operate.

Solution

Again, invest in the discovery phase — and find a top-notch software architect to create a blueprint for your solution architecture.

While it's recommended to start your IoT project small, focussing on essential solution components, you still need to have a bigger picture in the back of your mind.

Working with skilled cloud development and DevOps engineers is important, too. These specialists will help you devise a flexible, containerized solution that automatically adds computing resources in response to an increasing demand while reducing infrastructure costs.

Ignoring IoT Security Risks

From using data and networking protocols that fail to encrypt information to hard-coded device passwords and open security vulnerabilities in web and mobile applications, there are many security issues haunting IoT projects.

As a rule, they stem from poor project planning, wrong tech stack choices, and a failure to follow security best practices during IoT implementation.

Solution

Adopting the "security by design" approach is the only way to prevent data breaches, DDoS attacks, and other cybersecurity risks.

This approach encompasses a variety of activities, from securing connected solutions at the networking layer to configuring a reliable cloud infrastructure for issuing and installing over-the-air (OTA) firmware updates.

Most importantly, IoT security requires a convergence of your operational and information technology teams, as well as employee onboarding and education.

And all these things should be taken into account during project planning.