Picture a mid-sized manufacturer that wants to attach sensors to its production-line equipment, monitor downtime in near real time, and consolidate all that information in a single system. The project rationale is clear, the budget has been approved, and the technical team is ready. Yet one question remains unresolved: which platform? Getting that answer wrong can mean a costly system overhaul five years down the road or a permanent drag of high integration expenses. Selecting an IoT platform is no different from selecting an ERP — and in some respects it is considerably more complex.
The Internet of Things, at its core, refers to the infrastructure through which physical devices collect and transmit data across a network. But that definition says nothing about what a platform must actually do. Solutions on the market today cluster around three distinct axes: device-centric platforms, data-centric platforms, and ecosystem-centric platforms. These approaches are not mutually exclusive, but their priorities, cost structures, and long-term lock-in risks diverge sharply enough that a manager must understand the differences before committing to any one direction.
Device-centric platforms focus on connectivity and hardware lifecycle management — provisioning sensors, pushing firmware updates, tracking faults, and monitoring device health across a fleet. For industrial environments running hundreds or thousands of sensors, this focus is entirely rational. The weakness is that analytics capability tends to be thin; making sense of the collected data usually requires a separate toolset or an additional software layer. When calculating total cost of ownership (TCO), the licensing and integration cost of that extra layer must be included from the outset, not discovered after go-live.
Data-centric platforms invert that priority. Device management is treated as a prerequisite rather than a core competency; the real strength lies in ingestion, storage, reporting, and visualization. These platforms can consolidate streams from heterogeneous sources into a single repository and surface meaningful operational indicators. They suit firms that already have some device infrastructure in place and now want decision-support from that data. The risk is protocol compatibility: when a device uses a non-standard communication protocol, integration costs can escalate quickly. Reviewing the platform vendor’s supported protocol list — and its roadmap for expanding that list — before signing a contract is not optional.
Ecosystem-centric platforms are the broad-scope cloud infrastructures offered by large technology vendors. Device connectivity, data management, application development environments, and third-party integrations are bundled under a single umbrella. The appeal is obvious: rapid time-to-value and access to a wide developer community. The danger, however, is the highest lock-in risk of the three approaches. Once a firm is deeply embedded in an ecosystem, it becomes exposed to that vendor’s pricing revisions, service interruptions, and strategic pivots. An ROI analysis that ignores this ‘switching cost’ is incomplete.
For SMEs operating in Turkey, the practical constraint that shapes this decision is local technical support capacity. A firm that selects a foreign-headquartered ecosystem platform will often find itself navigating English-language documentation and support lines in a different time zone when problems arise. The number of local system integrators with genuine IoT depth is still limited, which makes independent consulting harder to source mid-project. Domestically developed solutions typically offer a narrower feature set, but they come with Turkish-language support, on-site implementation capability, and more flexible commercial terms — advantages that matter considerably for firms with lean technical teams.
Three questions should anchor the platform decision before any vendor shortlist is drawn up. First: is the primary priority device control, data analytics, or rapid application development? Second: if this platform needs to be replaced in five years, can data and processes be migrated without prohibitive cost? Third: what does the local support ecosystem look like today, and where is it realistically headed in two years? A manager who answers these questions clearly transforms platform selection from a technical preference into a strategic investment decision. The right IoT platform is not the one with the longest feature list — it is the one that fits the firm’s current operational maturity and leaves room to grow without trapping it in the process.
This article was originally written in Turkish by Gökhan MERCANOĞLU on June 24, 2013 and has been automatically translated into English and other languages using machine translation.