Stanley Black & Decker's Digital + AI Playbook: The Connected Jobsite

Most of the "AI in industry" coverage this year has been about software companies bolting a chatbot onto a workflow. Stanley Black & Decker (NYSE: SWK) is the more interesting case, because its products are physical and its AI is unglamorous: it lives in a concrete sensor predicting compressive strength, a fleet-telematics network that finds a missing nail gun, and an internal chatbot that mines product reviews so engineers can "test the next version of a product virtually." This company has been selling hammers for 180 years. Over the last five, it has quietly rebuilt itself around two ideas: the tool is no longer the product — the connected jobsite is, and AI is not a consumer feature but a high-stakes, governable engineering discipline. This post is an in-depth spotlight on SBD's digital offerings and, more specifically, on where AI is actually applied — inside its products and inside the company — and what that tells us about how the physical-industrial world is moving toward applied AI.

The Company in 2026

Stanley Black & Decker is a global tools-and-outdoor-solutions company headquartered in New Britain, Connecticut, with roughly 50,000 employees and a portfolio of 20+ brands: DEWALT, CRAFTSMAN, STANLEY, BLACK+DECKER, Cub Cadet, Hustler, Troy-Bilt, Osborn, and the Facom family, among others. It is the kind of name every contractor and every homeowner recognizes, which makes its strategic pivot the more notable part of the story.

The financial picture is a company in disciplined consolidation rather than hyper-growth: revenue is flat-to-down, but it is shrinking to focus. Selling the aerospace-fastening unit (CAM) is a clean signal of strategy — pruning an out-of-core business to sharpen the tools-and-jobsite center of gravity. Leadership is new at the top: Christopher Nelson is President & CEO (with former CEO Donald Allan Jr. now Executive Chair), and the technology organization is led by Chief Technology Officer Matt Griffiths, who has driven the data-and-technology transformation for over eight years.

The single most telling number for this post is not in the earnings release — it's a $1 billion U.S. investment announced August 12, 2026, with the company directing roughly half to research and development for "next-generation tools and breakthrough solutions," the other half to manufacturing capex, plus $60 million (through 2030) into DEWALT's "Grow the Trades" workforce program. CEO Chris Nelson framed it plainly: "Our U.S. investment strategy has multiple dimensions and goes far beyond expanding manufacturing — it's about igniting innovation, building world-class capabilities, and redefining the future of work in America." R&D at the head of a tools company's investment is the clearest institutional statement that the next competitive round is about technology, not just steel.

The Digital Pivot: From "Selling Tools" to "Total Jobsite Solutions"

The strategic reframe is worth naming explicitly, because it's the lens for everything below. For a century, SBD's value proposition was the product on the shelf. The current positioning — carried across the DEWALT site, its press, and executive commentary — is that DEWALT is "a leader in total jobsite solutions" and "a partner that can help lead customers towards the future of construction by tackling these challenges through a connected environment." That last phrase, from Brad Roberts, President of Construction Technology, is the whole strategy in one sentence: the unit of value is the connected environment, not the SKU.

The company has organized this under a formal Digital Accelerator — an Atlanta-based function it has run for years, whose mandate, per VP Robert W. Ross, "focuses on Artificial Intelligence, Robotics, and Emerging Technology," with the goal of "converting the exponentially increasing disruption brought by digital into a sustainable competitive advantage." That's the institutional home for the digital and AI work detailed below, and it predates the GenAI era — a fact that matters when we look at how deliberately (rather than reactively) SBD has approached the current wave.

The Digital Product Portfolio

The concrete expression of the pivot is DEWALT's Construction Technology suite — connected tools, software, and mobile applications, launched as a formal suite in May 2023 and now the backbone of the brand's "Jobsite Solutions" line. The pieces:

Two structural things stand out. First, the portfolio is explicitly cross-brand and interoperable: Tool Connect manages "DeWalt tools as well as other brands," and DEWALT's construction-tech page lists Autodesk, PypeServer, TigerStop, Siteaware, Trimble, and Procore as featured partners. SBD is not trying to own the whole software stack; it's positioning the tool as the sensor and actuator layer of an ecosystem the construction industry already uses. Second, the hardware and software are deliberately fused — the gateway, the sensors, and the tags are physical objects whose value is realized only in the software around them. That is the "total jobsite" idea made literal: you buy the tool, you get the data plane for free.

Where the AI Actually Lives in the Products

Zooming in on the AI inside the products (as opposed to the internal use, which comes next), it appears in two concrete places:

• Concrete (the most substantive product-AI). The Converge-powered concrete sensors "utilize AI to monitor and interpret real-time data" from DeWalt Signal Sensors. The "Concrete DNA" feature lets crews observe compressive-strength gain in real time and predict the critical strength milestone — so a contractor knows, to the hour, when a pour is strong enough to load, form, or strip. That is not a chatbot; it's a forecasting model embedded in a $ sensor, solving a genuinely expensive problem (a concrete pour that's loaded too early is a structural failure; one that's waited on too long is idle labor and crew). Curing milestone alerts, materials performance analytics, and BIM integration round it out. This is the archetypal "applied AI" product: narrow, high-value, and tied to a physical outcome with a dollar figure attached.
• Tool Connect Partner Network (the newest and biggest move — announced October 5, 2026). DEWALT launched what it calls "the construction industry's largest ecosystem" for jobsite asset tracking. The problem: crews lose visibility to tools the moment they leave a fixed gateway's range, and U.S. construction sites lose roughly $100M of tools a year (with total equipment theft running $300M–$1B and recovery rates under 25%, sometimes below 7%). The Partner Network turns infrastructure customers already own into detection nodes — no new hardware for existing Tool Connect users. Inaugural partners: Geotab (fleet telematics + "AI-powered insights" that report tool locations as they move across a vehicle fleet), Trackunit (off-road equipment telematics — machines detect tools on and around heavy equipment), and BoxLock (smart locks — a secured storage unit becomes a detection point that logs tool location when it's accessed or nearby). DEWALT's Eric Loeliger, Director of Product Management: "Our customers have told us that the network is the largest challenge in deploying tool tracking at scale… They get a single view of where their tools are, with significantly less effort. This is the broadest network of its kind in the industry." More partners — telematics, ERP, estimating — are expected within the year.

The Partner Network is the most important recent disclosure, because it reveals the business model of SBD's digital strategy: it is an ecosystem play, not a product play. DEWALT is building the detection standard (the Tool Connect chip and the Site Manager cloud) and then letting third-party platforms extend the mesh by reusing infrastructure that already exists in the field. The AI in that network (Geotab's "AI-powered insights") is a partner's, but it runs on DEWALT's asset graph. That's a quietly aggressive platform position: make the tool the reference node in a jobsite IoT network, and you own the data layer even when you don't own the intelligence layer. The 2026 AI Guidelines page DEWALT now publishes for its own use of AI signals the company is also formalizing the responsible-use side of these consumer-facing capabilities.

Inside the Company: The Applied-AI Program (the Better Story)

If the product-side AI is the story you can buy, the internal AI program is the story that teaches the lesson. CTO Matt Griffiths has spoken at length about how SBD has run its AI transformation, and the details are unusually specific about how a large, physical, consumer-facing manufacturer actually deploys AI at scale. The key facts:

• The strategy was "innovation, not restriction." When ChatGPT appeared in 2023, many companies locked down the technology. SBD did the opposite, for a very pragmatic reason: "Our theory was that [if we didn't] we were just going to drive all of our users underground on their personal devices and that we would lose them." They chose to build the sanctioned tool so the usage stayed inside guardrails rather than migrating to Shadow IT. That's a maturity signal — it treats adoption as an inevitability to be channeled, not a risk to be blocked.
• They built ChatSBD, an internal GenAI assistant, and it's a real product with real numbers. Starting with "low-hanging fruit" (IT and customer support), it is now in its third iteration, serving ~1,500 unique users daily and handling 15,000–18,000 prompts per day. For a 50,000-person company, that's meaningful, measurable internal adoption — not a pilot that died in a slide deck.
• The highest-value use case is mining unstructured data — specifically product reviews. SBD manages "over 80% of their data" as unstructured. For years it collected review text from Amazon, Home Depot, Lowe's, and Grainger but could only use star ratings. GenAI let it "analyze the text data, summarize it, gauge sentiment, and develop tools around these insights." Griffiths: "That allowed our engineers to test the next version of a product virtually, or for a quality and warranty team to understand why we were suddenly seeing an increase in a product being returned to our distribution centers." This is the killer example of applied AI in a physical company — it closes the loop between the customer on the job site and the engineer at the drawing board, using text that was always there but never usable.
• Griffiths now spends ~50% of his time on AI transforming IT itself. His framing of the frontier is explicit and agent-forward: "How does AI change the way we go about gathering requirements for IT projects? How does AI change the way in which the first version of code is done? The way that testing can be automated? Is there a world where 80% of your production issues are now being resolved autonomously using AI agents?… It's completely turning on its head the way that we've always thought about IT." In other words, the next phase of SBD's internal AI is not another chatbot — it's autonomous agents in the software factory. That's the same direction the Claude Code and harness posts on this site describe, coming from a tools company.

The "trustworthy AI" doctrine — the 80/20 rule

The most instructive part for anyone trying to apply AI in a physical, safety-relevant business is how SBD handles the trust problem. Griffiths is blunt: "It's really easy to build an AI chatbot that builds an itinerary for your vacation to Costa Rica. It's much harder to build an AI capability that consistently makes sure that when a customer asks which nails go into my framing nailer that you give the right answer. In that scenario, providing the wrong answer has a really serious implication and someone could get really hurt." That sentence — a wrong AI answer in a framing nailer can hurt someone — is the entire case for why industrial AI is a different discipline than consumer AI. His operating answer: SBD dedicates ~80% of its development time to the underlying platform (observability, explainability, transparency, bias detection, hallucination control) and only ~20% to the use case itself. Zero tolerance for GenAI giving incorrect information, enforced by running "varying large language models" behind a single trustworthy platform rather than bolting a model onto each feature.

Governance and adoption

Three practices, from Griffiths, that a company can copy directly: (1) an AI Governance Committee co-chaired by the CTO, CISO, and Chief Data Privacy Officer to define ethical use and policy — he calls it "in hindsight, the most important thing that we did," because the "AI blueprint" let the team build with business partners without needing constant approval. (2) Education through storytelling and demos, not PowerPoint: "GenAI is easier to understand when you see it working rather than trying to explain it on a PowerPoint slide." (The CISO, a risk-averse function, is now collaborating on seven GenAI use cases because Griffiths reframed AI as a cybersecurity asset.) (3) Don't wait to be "data-ready": "A lot of CDAOs get hung up on 'We need to get our data ready before we start introducing GenAI,' and that's absolutely not the case. In fact, in many cases, GenAI can help you fix your data issues." That last point is a genuinely counter-intuitive, practical correction to the standard "fix your data first" advice.

The earlier, "traditional AI" foundation

It's worth emphasizing that SBD was not a GenAI newcomer. Its AI history goes back years: it deployed H2O.ai's Driverless AI (with an "AI and Machine Learning Manager" building manufacturing-process models) well before ChatGPT, and its first CTO, Mark Maybury — an IEEE/AIAs fellow with a doctorate in AI from Cambridge — was describing the company's AI work (RPA through virtual assistants) in a MIT Sloan Management Review interview in March 2022. Griffiths is explicit that "the concept of using advanced models for building business capability was not new to us… we felt [GenAI] was a really important key in unlocking our ability to mine unstructured data." The GenAI wave was, for SBD, an extension of an existing machine-learning practice, not a reboot. That continuity is a major reason the program is credible rather than performative.

The Read: How a Physical Industrial Company Is Applying AI

Strip the specifics and SBD is the cleanest public example of a thesis this site keeps returning to — that in 2026, applied AI in the physical world is narrow, high-stakes, and governed, rather than broad and chatbot-shaped. Four patterns worth carrying into your own thinking:

The Honest Limits

• The physical products are still the revenue. Of a $15.1B business, the connected/AI offerings are a small, fast-growing sliver. There's no public figure for digital/services revenue, which is itself a point: the AI story is strategically important but not yet materially important to the P&L. Anyone claiming SBD is an "AI company" is well ahead of the numbers; "a tools company building a serious applied-AI capability" is accurate.
• The product AI is narrow by design — that's a strength and a ceiling. Concrete-curing prediction and tool location are excellent, but they are two problems. The broader "AI on the jobsite" vision (estimating, safety monitoring, scheduling) is mostly partner-driven or aspirational right now. SBD is a node in those ecosystems, not their builder.
• Most of the internal-AI detail is self-reported. ChatSBD's numbers (~1,500 daily users, 15–18K prompts) and the 80/20 effort split come from the CTO's own account. They're plausible and specific, but not independently audited. The "80% of production issues resolved autonomously by AI agents" line is explicitly posed as a question, not an achievement — it's the destination, not the status.
• Adoption ≠ impact at the enterprise. 1,500 daily users out of 50,000 employees is strong, but it's a fifth of the workforce. The "you'll never be AI-ready — just bite the bullet" stance is pragmatic, but it also means the program is continuously in flux, and governance for a physical-industrial context is genuinely harder to get right than for a software one.

Bottom Line

Stanley Black & Decker is the best current answer to the question this post set out to ask — how is a traditional, physical industry actually moving toward AI in a hands-on, applied way? The answer it gives is less theatrical than the software-world story but more instructive. The product side has AI in the place where it can move a physical outcome: a concrete sensor that predicts when a pour can be loaded, and a jobsite asset network that turns the fleet, the heavy equipment, and the smart locks into a shared detection mesh — a platform bet on owning the data layer of the connected jobsite. The internal side is where the lesson is sharpest: a deliberately built, governed, trustworthy-AI program (ChatSBD, an AI Governance Committee, an 80/20 split favoring the platform over the feature) that mines the 80% of its data that was always unstructured, and is now aiming squarely at autonomous agents in its own software factory. The through-line is discipline: narrow, high-stakes use cases, a hard line against wrong answers in a safety-relevant context, and the customer's own words as the flywheel. If 2026's applied-AI story is "the model is the commodity, the system around it is the product," SBD is the physical-industrial proof of the principle — a 180-year-old toolmaker, half a billion R&D dollars from now, betting that the next round of advantage is won not by a smarter hammer but by a connected, governed, and quietly intelligent jobsite.

References

• DEWALT — Tool Connect™ Partner Network press release (Oct 5, 2026; Geotab, Trackunit, BoxLock) [DeWalt Press]
• Stanley Black & Decker — Investing $1B in the U.S. to Drive Innovation (Aug 12, 2026) [IR]
• Stanley Black & Decker — 4Q and Full Year 2025 Results (Feb 4, 2026; CAM divestiture) [IR]
• DEWALT — Construction Technology Solutions (MSUITE, Tool Connect, Converge concrete sensors) [DeWalt]
• DEWALT — Introduces Suite of Construction Technology Solutions (May 23, 2023) [PRNewswire]
• Ohio Power Tool News — DeWalt AI-powered solutions (Concrete DNA / Signal Sensors) [Ohio Power Tool]
• Evanta (CDAO peer practice) — Matt Griffiths, CTO, on SBD's GenAI strategy (ChatSBD, 80/20 rule, governance) [Evanta]
• MIT Sloan Management Review — "Extreme Innovation With AI: Stanley Black & Decker's Mark Maybury" (Mar 2022) [MIT SMR]
• PRNewswire — H2O.ai Empowers Stanley Black & Decker (Driverless AI; Digital Accelerator) [PRNewswire]
• Stanley Black & Decker — About / Brands (portfolio) [SBD]
• Lawn & Landscape — Stanley Black & Decker names Christopher Nelson next CEO [Lawn & Landscape]

Financials from SBD's reported 4Q/FY2025 results (Feb 4, 2026); the $1B U.S. investment and Tool Connect Partner Network from company press releases (Aug 12 and Oct 5, 2026); product portfolio from DEWALT's construction-technology materials; internal AI program (ChatSBD usage, 80/20 effort split, governance) from CTO Matt Griffiths' own account via Evanta and prior MIT SMR coverage, as of October 28, 2026. Internal AI figures are self-reported by the company and not independently audited; no public figure exists for digital/services revenue.