Building a Better Display Local Device

We relied on a specific Onn model running a particular version of Android firmware. Our provisioning process required system-level customization, which meant an unexpected firmware update could potentially change or remove capabilities we depended on.

There was also no guarantee that the preferred model would remain available. Consumer electronics change quickly. A retailer can discontinue a product, introduce a revised model, or ship new firmware without considering how a specialized business has adapted that device.

The devices also carried someone else's identity. The case, onboarding screens, and startup sequence were designed for a consumer streaming product—not Display Local. That didn't prevent the player from working, but it made the experience feel less cohesive than we wanted.

The process probably looks different for a large company ordering thousands of units. As a scrappy startup, however, you have to approach it differently. You research, ask questions, order samples, solve unexpected problems, and learn as you go.

Because the Display Local application and mobile device management system were built around Android, we began by looking for Android media boxes. That still left us with a huge range of Android versions, chipsets, memory configurations, manufacturers, and firmware options.

Starting with the marketplaces

Our first stop was Alibaba and AliExpress. These marketplaces are useful for discovering products, but buying through a listing did not solve our long-term problem. Communication was often limited, product details could be unclear, and there was no certainty that a seller—or the exact configuration being advertised—would remain available.

Amazon presented a similar issue. Many generic Android boxes were available, but the source and long-term consistency of the hardware were difficult to establish. Prices were also often two or three times higher than the underlying hardware should cost.

Our real objective was not simply to find a box. It was to establish a repeatable hardware configuration and a relationship that gave us visibility into future changes.

Once we determined that marketplace listings were not a dependable long-term strategy, we began researching the manufacturers behind these devices.

Most manufacturers in this space are located in China. That made communication, sampling, and testing more involved. Language differences had to be navigated carefully, and international shipping made every sample significantly more expensive than its unit price suggested.

We identified several reputable manufacturers and began discussing chipsets, memory, storage, firmware, branding, minimum order quantities, and long-term availability. Over several months, we narrowed the field to two manufacturers we trusted and ultimately selected one partner for our initial production devices.

What mattered in that decision

• Clear, consistent communication

• Reasonable sample quantities

• Support for smaller initial orders

• Hardware consistency

• Competitive pricing

• Firmware expertise

• Customization capabilities

• Long-term product availability

We eventually focused our search on two Amlogic systems on a chip: the S905W2 and S905Y5. Our target configuration included 2 GB of RAM and 16 GB of storage.

There are less expensive chipsets and memory configurations, but our testing showed noticeable compromises. Video could lag, animations could become sticky, and the overall experience felt less polished. More expensive configurations performed well, but the improvements were not noticeable enough in everyday signage use to justify their additional cost.

We completed the firmware customization required for both chipsets, so the final decision came down to practical value. The S905W2 gave us the performance we needed without charging customers for specifications they would be unlikely to notice.

Selecting a processor and memory configuration was only the beginning. Specifications can tell you what a device should do. Testing tells you how it actually behaves.

Previously, preparing an Onn device for a customer took roughly 30 minutes. We had to work through the consumer onboarding process, connect a Google account, apply our settings, and complete the provisioning required to turn it into a signage player.

Our new device reduces that process to a minute or two. The factory configuration already reflects how the device will be used, so our team can prepare deployments faster and more consistently.

The physical device now carries Display Local branding, and that identity continues through the boot sequence and into the application. The result is not only faster to deploy; it feels like one complete product from the moment it is powered on.

Ordering a single sample from overseas can look almost ridiculous on paper. The device itself may be reasonably priced, but shipping, customization, and import costs can easily outweigh the cost of the hardware.

That changes as order quantities increase. When we order 10, 20, or 50 devices at a time, the fixed costs are spread across more units. The economics begin to make sense, and a customized device becomes competitive with—or more economical than—buying generic hardware through retail channels.

Shipping costs, tariffs, lead times, and minimum order quantities now form part of our planning. These details may never be visible to customers, but managing them carefully helps us maintain dependable supply and sensible pricing as Display Local grows.

Most customers will never need to think about firmware versions, manufacturing relationships, chipset comparisons, or international shipping. That is the point. We did this work so the device itself can fade into the background and simply run the content it is given.

We now have a direct manufacturing relationship and a defined hardware and firmware configuration. That gives us far greater consistency, along with better visibility into future component or firmware changes.

The S905W2 configuration gives us the balance we were looking for: dependable performance at an economical price. It also gives us a starting point for testing higher-performance chipsets, additional memory, and premium device options in the future.

This process took months of research, conversations, samples, testing, firmware work, and plenty of trial and error. But the outcome is more than a new Android box. It is a device selected and customized specifically for Display Local and the customers who depend on it.