Industry · Exploratory
Quantum Computing for Retail and Consumer Goods
Retailers and consumer goods firms depend on forecasting, recommendation, and optimization, and on formulating products from chemicals. Quantum computing has been proposed for each, but the best-known claim, an exponential speedup for recommendation systems, was undone by a classical algorithm.
Maturity
Exploratory; a few scheduling pilots and mostly research
Most credible quantum workload
Combinatorial optimization on annealers and hybrid solvers; simulation for product chemistry later
Hardware needed
Annealers for pilots; fault-tolerant machines for provable speedups
Strongest classical competition
Mixed-integer solvers, machine learning forecasting, and recommender systems
Instructive history
Quantum recommendation algorithm (2016) and its classical counterpart (2018)
Realistic horizon
Incremental optimization pilots now; clear advantage not established
Overview
Retail runs on thin margins and huge volumes. A supermarket chain manages hundreds of thousands of products across thousands of stores, schedules staff and deliveries, sets prices and promotions, and predicts what customers will buy. Consumer goods manufacturers formulate shampoos, foods, and cleaning products, and plan production and distribution on a global scale. Each of these activities generates optimization and data problems, which is why the sector is a frequent subject of quantum computing proposals.
The sector is also the home of one of the most instructive stories in quantum algorithms. In 2016 a quantum recommendation algorithm was announced with an exponential speedup, and in 2018 a classical algorithm inspired by it matched the speed under the same data-access assumptions. The episode taught researchers to treat claimed advantages in machine learning with caution, and it frames how to read quantum retail announcements today.
Why this industry is a candidate for quantum computing
Retail optimization problems, such as deciding which products to stock on limited shelf space, how to schedule workers, and how to route deliveries, are combinatorial and large. They are similar to the logistics and manufacturing problems for which annealing, QAOA, and quantum-inspired solvers have been tested, and a small percentage improvement across a large business is valuable.
Consumer goods firms also depend on chemistry. Flavors, fragrances, cosmetics, and cleaning products are mixtures of molecules whose properties come from quantum mechanics, and in the long term quantum simulation could help in the same way it may help the chemical industry, though at much lower priority than for catalysts or batteries.
Key facts and numbers
| Item | Detail |
|---|---|
| Quantum recommendation algorithm | Kerenidis and Prakash (2016) proposed an exponentially faster quantum recommendation system under data-access assumptions |
| Dequantization | Ewin Tang (2018) gave a classical algorithm with comparable speed, published at STOC 2019 |
| Grocery scheduling pilot | Save-On-Foods, part of Pattison Food Group, reported large reductions in optimization time with D-Wave hybrid solvers (2021) |
| Typical store assortment | A large supermarket stocks tens of thousands of distinct products |
| Annealer scale | D-Wave Advantage2 has more than 4,400 qubits, and embedding limits usable problem size |
| Grover-type speedup | Quadratic: about the square root of the number of options |
Where quantum computing could help
Recommendation systems and personalization
The problem: Predicting which products a customer will like from a large, sparse matrix of past purchases drives sales and customer satisfaction.
The quantum approach: The 2016 quantum recommendation algorithm used quantum linear algebra and a data structure to sample from low-rank approximations in time polylogarithmic in the matrix size.
Where things stand: In 2018 Ewin Tang showed that a classical algorithm, given comparable sampling access to the data, runs in similar time, which removed the exponential advantage. The episode launched a broader program of dequantization. Retailers use classical models such as matrix factorization and deep learning, which scale well.
Related algorithms: Quantum Principal Component Analysis, Quantum k-Means Clustering, HHL Algorithm
Assortment, shelf space, and pricing optimization
The problem: Choosing which products to stock, where to place them, and how to price and promote them involves large combinatorial choices with interactions between products.
The quantum approach: Problems are cast as quadratic binary optimization and given to annealers, QAOA, or quantum-inspired solvers.
Where things stand: Pilots with grocery and retail firms report improvements over existing heuristics, and a classical solver with the same formulation often does as well. The most reliable value so far comes from better modeling.
Related algorithms: Quantum Annealing, Quantum Approximate Optimization Algorithm (QAOA), Max-Cut with Quantum Algorithms (QAOA and Quantum Max-Cut)
Supply chain, inventory, and workforce scheduling
The problem: Retailers must schedule staff, replenish inventory, and route deliveries with demand that changes daily.
The quantum approach: These are scheduling and routing problems, and are attacked with hybrid solvers on annealers and with quantum-inspired hardware.
Where things stand: Save-On-Foods reported using D-Wave hybrid solvers for a scheduling problem and described large reductions in computation time. As with other pilots, the baseline and the classical share of the work matter, and the results are not proof of advantage.
Related algorithms: Quantum Annealing, Quantum Approaches to the Traveling Salesman Problem
Product formulation and materials
The problem: Developing flavors, fragrances, packaging materials, and cosmetics requires understanding molecular interactions and stability.
The quantum approach: Quantum chemistry simulation could assist when molecules are large or strongly correlated, and quantum machine learning has been proposed for property prediction.
Where things stand: Consumer goods firms are observers more than leaders in quantum chemistry. Most formulation relies on empirical testing and classical modeling, and quantum work is exploratory.
Related algorithms: Variational Quantum Eigensolver (VQE), Quantum Kernel Estimation
Who is working on it
- Pattison Food Group and its Save-On-Foods chain in Canada, which has worked with D-Wave on scheduling optimization.
- Large retailers and consumer goods firms that participate in quantum consortia or innovation labs, including Walmart's technology group and multinational manufacturers.
- D-Wave, IBM, and consulting firms that run optimization pilots for retail customers.
- Researchers who developed and then dequantized the recommendation system algorithm, including Iordanis Kerenidis, Anupam Prakash, and Ewin Tang.
- Quantum-inspired optimization vendors such as Fujitsu, Toshiba, and Hitachi, whose solvers have been used in retail and manufacturing settings.
Realistic timeline
| Horizon | What to expect |
|---|---|
| Now to 2028 | Hybrid and quantum-inspired optimization pilots, mostly in scheduling and assortment, benchmarked against classical solvers. |
| 2028 to 2032 | Early fault-tolerant machines enable small search and estimation demonstrations. Retail value remains incremental. |
| 2032 and later | If large machines arrive, quantum Monte Carlo and chemistry could reach consumer goods. Broad advantage in retail optimization is uncertain. |
How to judge quantum claims in this sector
Retail pilots usually report a percentage improvement or a speedup, so the first question is what it was compared with. Many retail systems still use rule-based or older heuristic methods, and almost any modern optimizer improves on them. A fair comparison uses a tuned commercial solver on the same data, and checks whether the quantum or hybrid result still wins.
The recommendation system history offers a second lesson. A quantum speedup that depends on a particular data-access model can disappear if a classical algorithm can use the same model. When a vendor claims an exponential advantage in machine learning, ask what assumptions about data loading and output it requires, and whether a classical sampling-based method has been tried under the same assumptions.
What stands in the way
- Strong classical tools: recommender systems and optimizers are mature, cheap, and getting faster.
- Data loading: retail data sets are large and sparse, and encoding them into quantum states is costly.
- Problem size: real assortment and scheduling problems exceed what current devices handle directly.
- Business case: gains must exceed the cost and complexity of new systems.
What organizations in this sector should do now
- Benchmark current optimization and forecasting tools against open-source and commercial solvers before running quantum pilots.
- Use quantum-inspired or classical hybrid solvers where they deliver measurable gains today.
- Treat machine learning claims skeptically, and ask about dequantization and data-access assumptions.
- Follow quantum chemistry developments if product formulation is a core capability.
Myths versus reality
- Myth: quantum computers will predict what shoppers want. Reality: the best-known quantum recommendation speedup was matched by a classical algorithm.
- Myth: pilot speedups prove advantage. Reality: gains often come from better formulation and classical components.
- Myth: retail is a leading quantum adopter. Reality: activity is modest compared with finance, pharma, and materials.
The bottom line
Retail and consumer goods offer many optimization problems and a good case study in how not to over-read machine learning claims. Companies can capture value today from better optimization and quantum-inspired solvers, and should follow quantum research with a critical eye.
Frequently asked questions
Can quantum computers improve product recommendations?
A 2016 algorithm claimed an exponential speedup, but a 2018 classical algorithm matched it under the same assumptions. No practical quantum advantage for recommendations has been shown.
Do retailers use quantum computers?
Some run pilots, such as a Canadian grocery chain using D-Wave hybrid solvers for scheduling, but production systems are classical.
What is dequantization?
The process of finding a classical algorithm that matches a claimed quantum speedup, usually by exploiting the same data-access assumptions.
Could quantum computing help design consumer products?
Possibly in the long term through chemistry simulation, but this is a low-priority, exploratory use.
What should a retailer do now?
Modernize optimization and forecasting, try quantum-inspired solvers, and benchmark any quantum pilot against strong classical baselines.
Keep exploring
- Algorithm: Quantum Principal Component Analysis
- Algorithm: Quantum k-Means Clustering
- Algorithm: Quantum Annealing
- Algorithm: Quantum Approximate Optimization Algorithm (QAOA)
- Hardware: D-Wave Advantage2
- Company: D-Wave Quantum
- Company: Multiverse Computing
- Research: Hardware-Aware Circuit Compilation Improves QAOA Performance on Real Devices
- Industry: Quantum Computing for Logistics
- Industry: Quantum Computing for Manufacturing
- Industry: Quantum Computing for Finance
This page is written for educational purposes and summarizes public announcements, published research, and published resource estimates (last reviewed September 30, 2026). Vendor and customer claims have not all been independently verified, timelines are estimates, and nothing here is investment or business advice.