+1 (978) 955-4300 · [email protected] EN · Fleet RFQ desk
Engineering journal

Picking Robot OEM vs Private Label: Four Ways to Buy Warehouse AMRs (And How to Tell Which One Fits You)


I run capital equipment procurement for a 900-person third-party logistics company. Warehouse automation has been my budget line since 2021 — roughly $6.2M a year, somewhere north of 20 vendors negotiated in that window, and a spreadsheet I maintain called AMR_TCO_v7.xlsx that I'm not proud of but will defend in a meeting.

Here's the thing I've learned about buying picking robots: there is no correct answer to "OEM or private label," or "direct or through a distributor." There are four routes. They fit different companies. The hardware conversation sounds roughly the same every time — payload, throughput per hour, charge cycle, integration protocol. The difference shows up in who owns the risk when something breaks at 2 a.m. on the Thursday before peak season.

The three variables that actually decide this

Before you compare quotes from any warehouse AMR supplier, get honest about three things about your own operation. Not the vendor's. Yours.

Site count and growth trajectory. One warehouse that will stay one warehouse is a different purchase from four warehouses with two more opening in 2026. This single variable reorders the whole decision. Multi-site operators need orchestration and fleet management across buildings; single-site operators mostly need somebody local who picks up the phone.

Who answers the pager. If your IT team is two people who also handle the WMS, the network, and the label printers, you cannot absorb a direct OEM relationship without help. Not a knock on your team — it's math.

CapEx or OpEx, and who signs. This is the one nobody mentions in the RFP and it kills more projects than technical fit does. A $2.1M capital request and a $48K/month subscription go to different people in your organization with different incentives.

Answer those three, and the route usually picks itself. Here's how each one plays out.

Scenario 1: Direct from the OEM

This is what most large operators picture when they think about Locus Robotics warehouse robots or any comparable collaborative AMR fleet — you contract with the manufacturer, they ship robots and software, you deploy.

Who this fits: multi-site operations, internal automation or industrial engineering capability, a WMS that's genuinely integration-ready, and a planning horizon past 18 months. FedEx's publicly announced AMR rollout with Locus in 2022 is the reference case everybody cites, and the reason it matters isn't the fleet size (go read the announcement if you want numbers). It's the deployment model — a standardized robot paired with centralized fleet orchestration, rolled across many buildings rather than customized per building. That's a multi-site architecture decision, not a single-purchase decision.

The cost structure: per-unit hardware plus a software layer — either licensed or subscription — plus integration. When people say "the robots cost X," they usually mean the hardware line only. The software line and the integration line are where the TCO spreadsheet gets interesting.

Hidden costs I've paid for: wireless coverage remediation (AMRs are unforgiving about roaming handoff gaps — a dead zone at aisle 14 shuts down a whole zone, not one robot), charging infrastructure and the electrical work to support it, spare parts inventory, and internal training hours that never appear on any quote.

The catch: direct means you are the first line of support. The OEM's field team travels. Your night shift doesn't. If you don't have someone who can read a log file and escalate intelligently, the SLA response time will feel theoretical.

Scenario 2: Through a warehouse robotics distributor or integrator

Same hardware, different counterparty. The warehouse robotics distributor takes the integration risk, handles the site survey, and stays on the hook for uptime.

Who this fits: single-site or two-site operations, thin IT, or a first automation project where you don't yet know what you don't know. Also fits companies where the WMS vendor is uncooperative — integrators are often better at dragging a legacy WMS vendor into an integration conversation than you are, because they do it every month.

The cost question to ask: is the hardware markup transparent, or is it bundled into a services figure? Both are legitimate. Only one lets you compare against a direct quote. Ask for the line items. If they won't break it out, that itself is information.

The question I now always ask: "How many brands do you carry, and how do you get paid on each?" Not hostile — just ask it plainly. A distributor recommending one brand out of eight because it's the best fit is a good partner. A distributor who carries one brand is not a distributor, they're a reseller, and their recommendation has exactly one possible answer.

The risk nobody quotes: vendor continuity. We had a regional integrator get acquired in 2023 and the new owner quietly exited the AMR business. Our contract survived. The relationship that made the contract work did not. If your integrator is under 40 people, ask what happens to your support agreement if they're acquired.

Scenario 3: Private label (the option that looks cheaper than it is)

This is the heart of the picking robot OEM vs private label question, and it's where I've seen the most money lost on the least drama.

Private label in this market usually means one of two things. Either a brand buys hardware from a manufacturer, puts their name on it, and adds their own software and service layer — or a robotics company sells a white-labeled version of its platform to a third party who resells it. Both are legitimate business models. Neither automatically means worse hardware.

What it does mean is that the layer that gets trimmed to hit the price is usually the expensive layer: on-site commissioning time, process redesign work, spare parts commitment, and long-term firmware investment.

Who it fits: honestly? Operations with strong internal engineering, a defined use case that isn't going to change, and the appetite to own long-term maintenance. That's a narrow band. Most companies who buy private label think they're in that band and aren't.

We spent two rounds of meetings in early 2024 talking past each other about the word "OEM." Our team meant "we buy direct from the manufacturer." Their team meant "we can manufacture to your specification and put your logo on it." Both are correct uses of the term. We discovered the mismatch when the first draft scope document arrived and it was a completely different project.

That's not a vendor failing. That's a vocabulary problem, and it's on the buyer to close it in writing before the second meeting.

The risk to price in explicitly: continuity. If a private label supplier exits in year three, your fleet doesn't stop working immediately — it just stops getting better, and stops getting parts. Ask directly: who manufactures the unit, what's their commitment horizon, and can you get a letter confirming parts availability for seven years. Any supplier with real confidence answers that in a day.

Scenario 4: Don't buy at all — subscribe

This is the one that goes against the instinct of every procurement person reading this, including mine five years ago. The default assumption is that buying beats renting on total cost, always. That's true often enough to be dangerous.

If your deployment horizon is under three years, or your order volume swings more than about 40% seasonally, or you genuinely don't know yet whether the process will work — subscription or robots-as-a-service transfers technology obsolescence risk and residual value risk to the supplier. That's not a small thing to hand over.

What I can tell you from our own numbers: across the deployments we've modeled, a five-year subscription ran 15–25% higher in total spend than an equivalent purchase. I don't have a clean public benchmark for this across the industry — every deal is structured differently and the published comparisons are usually vendor-produced. Treat that range as our data point, not a market fact.

So the honest framing: subscribe when uncertainty is the dominant cost, buy when certainty is. If you know the volume, know the process, and will still be running it in year six, buying wins and the RaaS pitch is a cash-flow story dressed up as a strategy story. If you're piloting a process change in a category you've never automated, paying a premium for the option to walk away is a legitimate purchase.

How to tell which one you are

Four questions. Answer them in writing, separately, before you talk to any vendor.

  1. How many sites in 2028, and how confident are you? Two or more with high confidence → lean direct. One, or uncertain → lean distributor.
  2. Name the person who takes the 2 a.m. call. If you can't name one, you're not ready for direct, and that's fine — it's a staffing answer, not a failure.
  3. Does your WMS vendor return integration calls? If no, a distributor or integrator buys you leverage you don't currently have.
  4. Has anyone in your company automated a warehouse process before? If this is the first one, don't buy private label and don't buy direct. Get a partner with skin in the game, learn the failure modes, then go direct on round two.

And whatever route you choose, put one line in the RFP that most buyers forget: confirm compliance with ISO 3691-4 for driverless industrial trucks, and if you're in the US, ask about ANSI/ITSDF B56.5 and UL 3100 for the platform. Not because your legal team will ask — because a supplier who can answer immediately has done enterprise deployments before, and one who has to "check with the team" tells you something about where you'd sit in their priority list.

Where Locus fits, and where it doesn't

I'll say the honest part. Locus Robotics is built for multi-site enterprise operations with real order volume and a willingness to change warehouse process — collaborative AMRs that work alongside people, coordinated through LocusONE orchestration. If you're running a 60-person single-site 3PL doing 300 orders a day, this is probably more platform than your operation justifies right now, and you should be looking at a smaller integrator-led deployment or a subscription pilot. That's not a product flaw. That's scope.

The same logic applies to any large AMR purchase. If your warehouse still routes pickers by memory and your racking layout hasn't changed since 2016, no robot will fix the underlying process — and you'll spend $1.8M discovering that. Fix the process first. Then buy the fleet.

The cheapest quote is a number. The right route is a decision about who absorbs risk. Those are different questions, and only one of them belongs in a spreadsheet.

Jane Smith

About Jane Smith

I’m Jane Smith, a senior content writer with over 15 years of experience in the packaging and printing industry. I specialize in writing about the latest trends, technologies, and best practices in packaging design, sustainability, and printing techniques. My goal is to help businesses understand complex printing processes and design solutions that enhance both product packaging and brand visibility.

Leave a technical question