There are three ways to get a robot into your building. They differ in price, and more importantly they differ in who absorbs the loss if the deployment does not work — which for a first project is the more relevant question.
Buying new
Advantages. Lowest total cost if the deployment succeeds, since no financing or rental margin is paid. Full warranty and support. Known condition and history. Freedom to modify, relocate or resell. And robot arms in particular retain value reasonably well.
Disadvantages. The largest cash outlay at the point of greatest uncertainty. Capital committed before the deployment has proven itself. All risk sits with you.
Suits. Organisations with available capital, high confidence in the application, a long expected life, and a reason to need modification freedom.
What to verify. Warranty duration and exclusions — specifically whether the battery is covered. Where the service centre is and what the response time is. Which spares are held locally and with what lead time. And who supports the software.
The framing. Buying new is cheapest if it works. The alternatives cost more precisely because they transfer the risk of it not working, so the comparison is really about how confident you are.
Renting
The most underused option and frequently the right one for a first deployment.
Advantages. Little or no capital. Maintenance and availability risk sit with the provider. You can stop if it does not suit. Access to current models rather than whatever you bought three years ago. And it answers the question with real experience rather than analysis.
Disadvantages. Higher total cost over years. Limited customisation. Dependence on the provider. And the equipment is not an asset.
Suits. Seasonal or event-driven needs. Organisations without a technical caretaker. Anyone uncertain whether the application will work. And trials before committing.
The threshold to buy. When accumulated rental approaches purchase price and confidence in the application is established.
What to check in a rental contract. Whether content preparation is included, support response time, liability for damage, whether upgrading to a newer model is possible, and the termination terms.
The strongest argument. On a first project, the uncertainty is not whether the equipment works but whether your organisation can operate it. Rental answers that at low cost.
Buying used
A real option requiring a clear-eyed assessment.
What is worth buying. The mechanical structure. A well-maintained arm or base has a long working life, and the price difference against new is substantial.
What carries risk. Battery condition, which is the largest and hardest to verify. Controller obsolescence and end of software support. No warranty. Programming tools that may not run on current computers. And configuration or account access that may not transfer.
What to check before buying. Operating hours. Measured repeatability rather than claimed. Joint play and gearbox condition. Whether the controller is still supported. Whether spares are still available. And a full run from charged to depleted to measure real battery capacity.
What to add to the price. Likely battery replacement, possible controller upgrade, and reloading all content and configuration from scratch.
The conclusion. Sensible where you have the technical capability to assess and maintain equipment. High risk where this is a first project and you do not know what to check.
Comparing by risk rather than price
A more useful frame than the price comparison.
Buying new. You carry all the risk that the application does not work. In exchange, the lowest cost if it does.
Buying used. You carry both the application risk and the equipment condition risk. The highest total risk of the three.
Renting. Most risk transfers to the provider, and you pay for that transfer through the rental margin.
How to choose. The more certain the application, the more buying makes sense. The less certain, the more renting does. And used only where you can genuinely evaluate condition.
The observation about first projects. On a first deployment the largest uncertainty is organisational rather than technical — whether the caretaker role gets filled, whether content gets maintained, whether staff cooperate. Rental is the cheapest way to test all three.
Later projects, with those questions answered, favour purchase.
Questions to ask in every case
Applicable regardless of acquisition route.
How many of these are running locally, and where? Ask to visit one.
Who repairs it, where, and within what response time? The most important question and typically asked last.
Which spares are held locally, and what are the lead times?
Who loads content, how long does it take, and is it chargeable?
What is the real operating time? With screen bright, audio on and movement happening.
What does a battery replacement cost?
If your company stops trading, does the equipment still function? An uncomfortable question, and essential for anything depending on a cloud service.
May we trial it in our own space? The single most informative question, because the reaction tells you more than the answer.
Frequently asked questions
Is buying new always cheapest?
Cheapest if the deployment succeeds. The alternatives cost more precisely because they transfer the risk of it not working, so the comparison is really about how confident you are in the application.
What is the strongest argument for renting first?
On a first project the largest uncertainty is organisational rather than technical — whether the caretaker role gets filled, content gets maintained and staff cooperate. Rental tests all three cheaply.
What is the biggest risk with a used robot?
Battery condition, which is the hardest thing to verify. Run it from charged to depleted to measure real capacity, and add likely replacement cost to the purchase price before comparing.
Which question is most informative when talking to any supplier?
Whether you may trial it in your own space. The reaction to the question reveals more than the answer — a confident supplier agrees under reasonable conditions.
More in Local ecosystem and Myths and questions.
