"Turnkey" is probably the most stretched word in automation quoting. In theory it means you hand over a requirement and get back a working cell that you can switch on and put into production. In practice it can mean anything from that, all the way down to a robot, a gripper and some guarding delivered to your door with a line in the small print saying the interface to your machine is "by others".
That is not usually anyone being dishonest. A robot cell pulls together several disciplines that have traditionally been sold separately, and each supplier tends to quote the part they are comfortable with. The problem is that you, the buyer, end up owning the gaps between them, and the gaps are exactly where cell projects go wrong.
This post sets out what a genuinely turnkey robot cell should include, where the scope most often stops short, and the questions worth asking before you sign anything.
Why the gaps matter more than the parts
Take a fairly ordinary machine tending cell. Somebody designs the base frame and guarding, somebody designs the gripper, somebody writes the robot program, somebody writes the PLC logic, somebody designs and builds the control panel, and somebody works out the safety concept and validates it. On a lot of projects those are four or five different organisations.
Each of them can do their own part well and the cell can still be late, because nobody owned the joins. The gripper gets designed before anyone has agreed how the robot will know a part is actually in it. The guarding gets drawn before the safety concept is settled, so the light curtain ends up bolted on in a place that makes loading awkward. The handshake with the machine tool is assumed to be simple until someone on site discovers the machine needs an extra option enabled before it will talk to a robot at all.
When a robot cell runs late, the cause is usually the same. Every supplier delivered their piece, and nobody was responsible for the cell.
A turnkey supplier's real job is to own those joins. If a quote breaks the cell into packages and quietly hands some of them back to you, it is not turnkey, whatever the cover page says.
What a complete cell should include
Every cell is different, but the scope of a properly turnkey project looks broadly the same each time. This is the list we would expect to see covered, along with the wording that tends to signal something has been left out.
| Scope item | What good looks like | Watch for |
|---|---|---|
| Requirements and cycle time | A written specification agreed up front, with the cycle time proven against real robot motion rather than estimated | "Cycle time to be confirmed during commissioning" |
| Mechanical design | Layout, base frames, guarding, infeed and outfeed, fixtures and robot mounting, designed in 3D alongside the controls | "Guarding to customer's standard" or "layout by customer" |
| End of arm tooling | A gripper designed around your actual parts and every variant you run, with part presence sensing | "Standard gripper" with no mention of your part range |
| Robot and PLC programming | Robot programs, PLC logic and HMI written by the team that designed the cell, including fault recovery | Programming priced by the day with no defined end point |
| Electrical design and panel | Full schematics, panel design and build, and field wiring to suit the cell as designed | "Power supply to panel by others" with no detail on what that means |
| Interface to your equipment | The signals, doors, clamps and handshakes with your machine or line clearly defined and included | "Machine interface by others" |
| Safety and compliance | Risk assessment, safety system design and validation, and the declaration of conformity for the finished cell | No mention of who signs the declaration |
| Testing and commissioning | Factory acceptance test before shipping, then installation, on-site commissioning and a site acceptance test | FAT listed as "optional" or missing entirely |
| Handover | Operator and maintenance training, drawings, program backups, manuals and a recommended spares list | "Documentation in supplier's standard format" with no list |
None of those lines are exotic. The point is that every one of them has to sit with somebody, and if the quote does not say it sits with the supplier, the safest assumption is that it sits with you.
The safety paperwork is part of the cell
This is the item most often treated as an afterthought, and it is the one with real legal weight behind it.
An industrial robot on its own is classed as partly completed machinery. It arrives with a declaration of incorporation and assembly instructions rather than a declaration of conformity, because it cannot do anything safely until it has been built into something. Once the robot, the guarding, the tooling and the machine it serves are linked together to do a job, the cell as a whole has to be assessed, given a declaration of conformity and UKCA or CE marked by whoever is acting as its manufacturer. That is often the integrator, but it can just as easily end up being the end user if nobody else has taken it on.
So it is worth asking directly who will carry out the risk assessment, who will validate the safety functions, and whose name will be on the declaration. If the answer is vague, that is a cost and a responsibility that has not been priced.
Worth knowing: ISO 10218-2, the standard that covers integrating robots into applications and cells, was revised in 2025 for the first time since 2011, and it now takes in most of the collaborative robot guidance that used to sit in ISO/TS 15066. If you are buying a cell now, it is reasonable to ask which edition it is being designed to. And if the cell is heading to Northern Ireland or the EU, the new EU Machinery Regulation applies from 20 January 2027, so check your supplier is working to it.
Interfaces are where turnkey quietly stops
The robot side of a cell is the part everybody is comfortable quoting. The connection to your existing equipment is where scope tends to get fuzzy, and it is also where some of the most expensive surprises come from.
Machine tending
For a CNC machine tending cell, the robot has to know when the machine has finished, open or wait for the door, release and clamp the workholding, and tell the machine to start the next cycle. Depending on the machine, that might need an automatic door, powered workholding and a robot interface option from the machine tool builder. Those can carry a real cost, and if nobody has checked what your particular machine supports, you find out at the worst possible moment.
Palletising and end of line
For a palletising cell, the questions are about what feeds it and what takes product away. Who supplies the infeed conveyor? How do empty pallets arrive, and how do full ones leave? Does the cell need to talk to a labeller, a stretch wrapper or your warehouse system? Each of those is a boundary, and each boundary needs an owner.
A good turnkey quote will name every interface, say what signals pass across it, and say who is responsible for the equipment on the other side. If the interface list is missing, ask for one before you compare prices, because two quotes that look similar on cost can be miles apart on what they actually cover.
Looking at a robot cell?
We design, build, programme and commission complete cells, so one team owns the mechanical design, controls, safety and handover from first sketch to running production.
Robot Cells →Standard or bespoke?
Not every cell needs to start from a blank sheet. Machine tending and palletising cells share the same core layout far more often than they differ, so a supplier with a proven standard design can give you a clearer scope and a faster quote, then adapt the details to your parts and floor space. That is the thinking behind our standard robot cells.
Larger, multi-robot or heavily bespoke systems are a different proposition. They usually involve specialist mechanical partners, and that is fine, as long as the contract still gives you one party who is accountable for the whole result. The test is simple. If something goes wrong between the mechanical build and the controls, do you have to referee it, or does your supplier?
Questions to ask before you sign
If you are comparing quotes for a robot cell, these questions will tell you more than the headline price:
- What is excluded? Ask for the exclusions in writing, not just the inclusions
- Who owns the interface to my machine or line, and has anyone checked what my equipment actually supports?
- Who carries out the risk assessment and signs the declaration of conformity for the finished cell?
- How is the cycle time being proven? Simulation against real robot motion is far more reliable than an estimate
- What happens at FAT, and what has to pass before the cell is allowed to ship?
- How will the cell recover from a fault, such as a stop mid-cycle or a part not arriving where it should?
- What do I get at handover? Drawings, program backups, manuals, training and a spares list should all be named
- Who do I call when it stops, and is that the same team that built it?
A supplier who answers all of those clearly has probably thought the project through. One who struggles with the safety and interface questions in particular is telling you where the problems are likely to be.
How we approach it
Duke started as a controls and robot programming business, which shapes how we deliver cells. We design the cell around how it will actually run, and the same team handles the mechanical design, tooling, robot and PLC programming, electrical design, panel build, safety validation and commissioning. Programs and logic are proven in simulation before anything reaches your site, fault recovery included, so site time is spent commissioning rather than debugging.
Whether it is a standard machine tending or palletising cell or a larger bespoke system, you get one contract, one point of contact and one team responsible for the result. If you are scoping a cell and want a second opinion on what should be in it, we are always happy to talk it through.