Retail (Robotics / Shelf Intelligence) · Deep dive

Simbe Robotics

The eleven-year-old South San Francisco robotics company selling Tally — a five-foot, LiDAR-and-RealSense shelf-scanning robot — into Wakefern, Schnucks, BJ's, SpartanNash, Albertsons and CarrefourSA on a Robotics-as-a-Service subscription reportedly starting near $2–4k per store per month, with $104M raised across three rounds and Goldman Sachs' Growth Equity arm underwriting the Series C at a more-than-tripled valuation.

emerging

The question that decides it: **Does Tally recover more incremental sales from out-of-stocks than a $30–50k-per-store low-wage in-store labor pool can — measured as a repeat-purchase rate above 80% at three years after first install and unit economics that hold up after Walmart–Focal Systems, Ahold Delhaize–Badger Marty, and shelf-edge camera systems make vision-only shelf audit an in-house feature?** The answer is testable across the 2027 renewal window on Simbe's earliest chain-wide deployments (Schnucks 2017, BJ's 2020, Wakefern 2021): count the retailers that expand rather than churn, and compare Tally's per-store $216K sales-lift claim against the fully-loaded cost of a store-clerk audit shift or a fixed-camera network in the same aisle.

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Simbe Robotics is the eleven-year-old South San Francisco company behind Tally, an autonomous five-foot shelf-scanning robot deployed at Schnucks, Wakefern (ShopRite), BJ’s Wholesale, SpartanNash, Albertsons, Giant Eagle, CarrefourSA and Decathlon USA. Simbe has raised $104M across a $26M Venrock-led Series A (2019), $28M Eclipse-led Series B (2023) and $50M Goldman Sachs Alternatives-led Series C (Oct 2024). Tally 4.0, unveiled Jan 12, 2026, adds twelve-hour runtime and 360-degree cameras. Simbe claims ~1,000 stores in ten countries, 600M shelf gaps detected and 80M promotion errors caught in ten years. The question is whether that footprint compounds — or shelf intelligence migrates to Walmart-owned Focal Systems’ fixed cameras faster than Simbe converts pilots into chain-wide contracts.

Founding story

Simbe emerged from Willow Garage. The Menlo Park robotics incubator that seeded ROS, PR2 and the founders of Fetch, Savioke and Suitable Technologies wound down in 2013–2014. Mirza Shah — a Penn State CS graduate who had spent 2012–2013 at Willow Garage contributing to the ROS core — was one of the scattered engineers. Brad Bogolea, a fellow Penn State alumnus with a decade in wireless-sensor and instrumentation work for the power industry, was hunting for a physical-world data problem paired to a labour-intensive incumbent process. The anecdote he repeats (Cherubic, Grocery Dive, Robohub) is walking into a grocery store, finding the shelf empty, and concluding on-shelf-availability was a data problem human labor economics could not solve at the required frequency. Bogolea, Shah and industrial designer Jeff Gee — another Willow Garage / Suitable Technologies alum — incorporated Simbe in 2014 and launched Tally in November 2015, the first commercially deployed autonomous shelf-audit robot.

How it works

Tally is a battery-powered autonomous mobile robot roughly five feet tall on a differential-drive base. Its sensor stack, per Simbe and the OLogic manufacturing spec, is a dozen-plus fused sensors: 2D RGB cameras with autofocus and HDR; Intel RealSense 3D depth cameras (upgraded to a RealSense LiDAR camera in Tally 3.0); dedicated navigation LiDAR; and an embedded Nvidia Jetson (broadened to a full Nvidia stack in Tally 4.0) doing edge inference. It navigates autonomously around shoppers and staff, traversing each aisle multiple times a day — Simbe cites 15,000 to 30,000 SKU inspections per hour. Onboard OCR reads price and promo tags; barcode inference matches shelf position to planogram; proprietary CV models detect out-of-stocks, low-stocks, misplaced items, mispriced tags and planogram gaps. The AWS-hosted back-end aggregates imagery into a store dashboard, exposes an API to retailer inventory and merchandising systems, and — as of Tally 4.0 — publishes a walk-through digital twin from the dual fisheye cameras. Simbe claims 99%+ shelf-condition accuracy and ten-times the out-of-stock catch-rate of manual audits.

Product and business overview

Three product surfaces. Tally is the roving robot — the flagship data-capture platform sold as-a-service. Simbe Vision is the software layer: store-intelligence dashboards, alerting, planogram compliance and API integrations into retailer ERP, replenishment and workforce systems. Simbe for Merchants — announced November 2025 — repackages Tally’s anonymised shelf data as an analytics feed sold to CPG brands (Coca-Cola, Kraft-Heinz, P&G class of buyer) hungry for real-time out-of-stock and share-of-shelf data on their SKUs. That third motion opens a second revenue pool (brand-side ad and data budgets) sitting on top of the retailer-side subscription. Ancillary 2025 additions include Tally Spot (a fixed camera for coolers and promotional endcaps) and fresh-department capabilities that let Tally audit produce and bakery — previously the hardest aisles for a vision-first robot.

Business model and pricing

Simbe books revenue as a Robotics-as-a-Service subscription: one bundled monthly fee covering hardware, on-site maintenance, the software platform and cloud — zero upfront capex to the retailer. Simbe’s FAQ confirms RaaS without listing prices; independent reporting (Cherubic Ventures, Robotomated, PricingNow) triangulates a per-store monthly fee historically in the $2,000–$4,000 band, indexed to store size, SKU count, scanning frequency and whether RFID is enabled alongside computer vision. Contracts are multi-year with pilot-to-chain expansion — Schnucks piloted three stores in 2017 and has since brought Tally into more than half its footprint; Wakefern piloted in 2020 and expanded across ShopRite banners. Simbe for Merchants (2025) adds a CPG-side revenue line that scales with existing retailer deployments rather than requiring new robot fleet growth. November 2025 statements disclose Simbe tripled contracted ARR in 2023 and roughly doubled store footprint in 2024.

Traction over time

• Nov 2015: Tally 1.0 commercially launched — first autonomous shelf-audit robot.
• 2017: Schnucks pilots Tally in three St. Louis stores (Retail Dive).
• Sep 2019: $26M Series A led by Venrock; SoftBank Robotics adds inventory financing.
• 2020: BJ’s and Giant Eagle sign multi-store deployments; Tally 3.0 ships with RealSense LiDAR and Nvidia Jetson edge compute.
• 2021: Wakefern (ShopRite) begins Tally pilot; Save Mart and Hy-Vee follow.
• Jul 2023: $28M Series B led by Eclipse; Simbe reports tripled contracted ARR for the year.
• Summer 2024: CarrefourSA (Turkey) launches Tally in select stores.
• Oct 24, 2024: $50M Series C led by Goldman Sachs Alternatives; valuation more than tripled off Series B.
• Nov 2025: Ten-year milestone — 600M shelf gaps detected, 80M promotion errors fixed, 18B price tags scanned, ~1,000 stores in ten countries.
• Jan 12, 2026: Tally 4.0 unveiled; ships mid-2026.

Market analysis

Two addressable pools. The narrower is the shelf-scan subscription itself: ~40,000 US supermarkets plus ~40,000 mass-merchant, warehouse-club and dollar-store units, at $2–4k per store per month, implies a US retail-ops TAM in the $2–4B annual range at full penetration — realistically $500M–$1B accessible after stripping Walmart (Focal-committed), Ahold USA (Badger-committed) and price-sensitive small independents. The wider pool is CPG-side store intelligence: IRI, NielsenIQ, Circana and RetailNext already sell on-shelf availability and share-of-shelf data at multi-billion-dollar scale using panel and POS methods; Simbe’s argument is that continuous vision-based ground truth beats panel-and-POS on freshness and share-of-shelf accuracy. Structural forces favor Simbe: post-2021 US retail wage inflation raised a store shelf-audit clerk shift to $30–50k per year, and a Coresight-Simbe February 2026 study estimates $137B in annual US out-of-stock losses.

Competitive intel

Focal Systems is the strategically loudest competitor — Walmart acquired it in 2022 and has been rolling out its fixed shelf-edge cameras across 400+ US and Canadian stores. Focal’s wedge is not a better robot but no robot at all: a shelf-mounted camera per bay, cheaper per unit, unobtrusive to shoppers, embedded in Walmart’s IT stack. That validates the willingness to pay for vision-based shelf audit while removing the largest retailer from Simbe’s TAM. Bossa Nova is the cautionary parallel — Walmart terminated its 500-store contract in November 2020, Bossa Nova cut 61 staff and folded its European unit per TechCrunch. Badger Technologies, owned by Jabil (~$12–15B contract manufacturer), fields Marty at ~500 Stop & Shop and Giant Food stores under Ahold Delhaize USA — a lock on the second-largest US grocer group. Pensa Systems (drone-based, CPG-focused) is the direct competitor for Simbe for Merchants revenue. Trax Retail — capitalised above $1B on paper — has retrenched toward CPG analytics and lost retailer-side momentum through 2023–2024. Zebra Technologies uses its Fetch Robotics acquisition to bundle mobile-robot shelf audit under its hardware relationship. Grabango wound down US operations in 2024 and Standard Cognition folded into Alimentation Couche-Tard’s Circle K in 2023.

History and evolution

2014–2015: incorporation and Tally 1.0 launch make Simbe the first mover. 2017–2019: Schnucks pilot, then a $26M Series A with SoftBank Robotics as strategic partner. November 2020: the near-death moment — Walmart cancels Bossa Nova and the category takes a reputation hit; Simbe survives because it never bet on Walmart. 2021–2023: methodical wins at Wakefern, Save Mart and Hy-Vee against a capital-scarce robotics backdrop; Eclipse leads the $28M Series B in July 2023 without a hero valuation. October 2024: Goldman Sachs Alternatives’ Series C at a more-than-tripled valuation signals Simbe survived the shakeout. 2025: Simbe for Merchants launches; Tally Spot introduces a fixed-camera SKU into what had been a pure roving-robot company. January 2026: Tally 4.0.

What people say

The case for. Trade press is favorable — Retail Dive, Grocery Dive, Progressive Grocer and The Shelby Report profiled Tally repeatedly through 2024–2026. Simbe’s case-study claim of $216K per store in annual sales lift from improved on-shelf availability is the most-quoted number in the category. Glassdoor is strong (4.7 of 5 across 12 reviews mid-2026), citing collaborative leadership. Goldman Sachs’ October 2024 investment cites a tripled prior valuation.

The complaints. The absence of a Walmart contract is a persistent gap now that Walmart owns Focal; Ahold’s US banners are locked into Badger Marty. Packaging Europe reports skepticism from stock-control workers who say the robot mishandles top and bottom shelves and mixed SKUs. Reddit surfaces staff friction — clerks having to walk over and dismiss Tally’s false-positive gap alerts — and shopper unease, though Simbe’s commissioned CarrefourSA study argues net-positive sentiment after six months. Selection bias in the $216K number is the biggest analytical risk: pilot stores were chosen for cooperative staff and higher pre-existing OOS rates, and no third party has audited the counter-factual. The Bossa Nova pattern — a headline pilot fleet at a top retailer terminated inside five years — is the reference-class overhang.

Outlook: the open question

The falsifiable question is whether Tally delivers more incremental margin than the two credible alternatives — a $30–50k-per-store shelf-clerk audit budget, or a fixed shelf-edge camera network like Focal’s — measured at the three-year renewal window on Simbe’s earliest chain-wide contracts. Answerable in 2026–2028. Schnucks (piloted 2017), BJ’s (2020), Wakefern (2021) and Giant Eagle (2020) all reach material renewals inside two years. If Tally’s sales-lift compounds and Simbe converts its ~1,000-store footprint into chain-wide contracts at Wakefern, BJ’s and Albertsons rather than seeing defection to fixed cameras, the RaaS thesis holds and Simbe for Merchants CPG-data begins to matter more than the robot. If a top-five renewal is downgraded, the Walmart-Focal precedent generalises and the category collapses toward embedded cameras. Intermediate case: Simbe survives in fresh, produce, bakery and floor-display audit — aisles fixed cameras struggle with — while dry-grocery migrates to fixed. In that world Simbe becomes a fresh-department and CPG-data business, with the roving robot as loss-leading data-capture. The 2027 renewal cohort will tell.

How to attack it

The specific wedge is CPG-brand-funded shelf intelligence that pays the retailer to install the sensors instead of charging them. Simbe today charges the retailer $2–4k per store per month for RaaS. Focal-inside-Walmart proved retailers will fund vision if it is capex-light and unobtrusive; the harder truth is that the biggest dollar buyer of on-shelf-availability data is the CPG brand, not the retailer. An attacker could invert Simbe’s pricing: give the fixed-camera network away free to mid-size regional grocers, book revenue from Kraft-Heinz, P&G, Coca-Cola, PepsiCo and Unilever paying for continuous share-of-shelf and out-of-stock data. IRI, NielsenIQ, Circana and Trax already sell that data pool at multi-billion-dollar scale.

Simbe’s exploitable weaknesses: (1) Form factor overhang: the roving robot has higher marginal cost, higher maintenance overhead and a shopper-visible failure mode than fixed cameras; per Focal’s public rollout numbers and Bossa Nova’s Walmart cancellation, this is where the market voted. (2) Retailer-pays economics: with wage inflation now easing off 2022 peaks and Amazon-style just-walk-out tech normalising, retailer willingness to pay $30–50k per store per year for shelf audit is retesting. (3) CPG channel underbuilt: Simbe for Merchants launched November 2025, five to ten years behind Trax and Pensa in CPG relationships. (4) No admitted top-five US retailer (Walmart, Amazon, Kroger, Costco, Ahold-US) is currently paying Simbe at chain-wide scale — a well-capitalised attacker with a Kroger or Costco anchor deal would flip the reference-customer narrative. (5) Hardware concentration risk: Tally 4.0 rides Nvidia Jetson and Intel RealSense; any supply disruption or price spike hits gross margin directly, unlike a pure-software competitor. (6) The Simbe for Merchants pivot puts Simbe in direct competition with its own retailer customers who guard first-party shelf data — the Amazon-versus-brand-seller conflict has to be managed and is not obviously solved.

Adjacent-segment play

The most attractive adjacent-segment is the CPG data business itself — Simbe for Merchants, expanded into a full IRI/Circana/NielsenIQ challenger — because the store fleet becomes a sensor network whose highest-value output is not what the retailer sees but what the brand pays for. That is a $10B-plus annual pool globally per NielsenIQ 2024 filings and Circana 2025 investor materials. Simbe would repackage Tally’s raw data as a syndicated feed sold to top-500 CPG brands: continuous, sub-daily out-of-stock rates by SKU by store, share-of-shelf trends by facing count, and promotion-execution audit tied to trade-spend budgets. Trax, Pensa and IRI-Panel are already in that market — but none of them owns the roving-robot ground truth that Simbe does.

A secondary adjacent is non-grocery physical retail: pharmacies (CVS, Walgreens, Boots), sporting goods (Decathlon already deployed), automotive parts (AutoZone, O’Reilly), home improvement (Ace Hardware) — retail verticals where SKU count is high, staff labor is thin, and out-of-stock cost per item is greater. Simbe already claims deployments in nearly a dozen sectors per November 2025 statements. A third is warehouse-adjacent cycle-count — Tally’s autonomous mobile base plus vision stack maps onto 3PL and wholesale-club backroom inventory audit, competing with Locus, Fetch (Zebra), 6 River Systems and Symbotic in a different but partially overlapping space. What does not generalise is the sub-100-store convenience-store segment (economics too thin per unit) and the sub-1,000-SKU specialty-boutique tier (human audit cheaper). A cleaner off-ramp is a data-licensing partnership with a Circana or NielsenIQ that keeps Simbe’s balance-sheet-light and monetises the sensor network without owning the CPG go-to-market itself.

Sources and further reading

• Simbe Raises $50 Million in Series C Funding Led by Growth Equity at Goldman Sachs Alternatives — Simbe Robotics, October 24, 2024.
• Simbe Raises $28M Series B, Led by Eclipse — Simbe Robotics, July 13, 2023.
• Simbe raises a $26M Series A for its retail inventory robot — TechCrunch, September 12, 2019.
• Simbe Marks 10 Years of Tally the Robot — PR Newswire, November 2025.
• Simbe Unveils Tally 4.0 — Simbe Robotics, January 12, 2026.
• Simbe’s Tally Robot Revolutionizing Grocery Retail With AI, Automation — The Shelby Report, March 5, 2025.
• Walmart ends five-year inventory robot contract with Bossa Nova Robotics — The Robot Report, November 2020.
• BIG NEWS: Walmart chooses Focal Systems Shelf Cameras for 400+ store rollout — François Chaubard / LinkedIn, September 2022.
• Shelf-scanning, AI-powered robots hit supermarket aisles to mixed reviews — Packaging Europe, 2024.
• How a Leading U.S. Grocer Improved On-Shelf Availability and Drove +$216K Annual Sales Per Store — Simbe Robotics case study.
• Simbe Robotics — Customer Story — Foxglove.
• Simbe Tally 3.0 Review & ROI Calculator — Robotomated, 2026.