How To Maximize Competitive Advantage

Competitive advantage is becoming much more fluid as market positions shift faster across industries, with advantages eroding and differentiation narrowing.

Despite this increased volatility, companies often fail to actively monitor shifts in their competitive edge. Join Matt Banholzer and Laura LaBerge on Inside the Strategy Room as they discuss their latest research on the erosion of competitive advantage and share what companies can do to maximize their edge over peers as AI accelerates change.

TL;DR

A ~48-minute episode of Inside the Strategy Room on the McKinsey & Company channel, published 10 August 2026 — host Sean Brown with Matt Banholzer (senior partner, Chicago; global leader of the strategic growth and innovation practice) and Laura LaBerge (senior expert, Connecticut), on new research into the erosion of competitive advantage.

1. The two framing statistics. “As few as 10% of the organizations they recently surveyed have full alignment on what their competitive advantage actually is.” And: “More than 60% of industries over the past decade have seen an 11% increase in their shuffle rate” — a measure of how quickly market leaders and laggards change places. LaBerge adds a companion finding from earlier in the year: two-thirds of companies track performance variance only at the business-unit level, “so they don’t even see the variance to a precise enough degree that they can do something about it.”
2. The definition, with three load-bearing qualifiers. Competitive advantage is “the distinct set of hard-to-replicate assets and operating models that a company creates that earn superior returns over time.” Banholzer unpacks each: hard to replicate — “if everyone can do what you do, that’s the beauty of capitalism… it’s just par”; superior returns — “if it’s something you do really well, but the market isn’t paying for it, nobody really cares… unique is not the same as hard-to-replicate superior returns”; and over time — durability “measured not in months or even a couple of years, but in decade-plus.”
3. Why now. Banholzer’s answer is not that the environment is uniquely turbulent (“you can always pick your favourite thing that is changing the world”) but that the measurement has become possible: two decades of research correlated shareholder return to things visible on a balance sheet, and “now is the time to actually look at what are the root causes upstream of that” — the qualitative, previously unmeasurable claims like “culture is an advantage” or “our intellectual property is an advantage.” His stated bar: “what we don’t want to do is just make this another exercise that creates yet one more framework.”
4. Seven sources of advantage, detectable from outside. LaBerge ran the top 5,000 global companies against three tests: is there variation within an industry (otherwise the comparison is useless), are the markers specific enough to act on, and are they right. Metals and mining was chosen as the hard case — “commodity, B2B, not an industry where the majority of players are household names” — and across 180+ companies the markers spread widely rather than clustering, some specific “to the level of this particular metal, this particular mine,” others broad (“do we have government relationships”). They resolved into seven themes that held across industries: innovation and IP, customer access or channel, brand and reputation, unique assets and resources, operational excellence, scale, and financial strength. Her honest scoping of accuracy: where the answers weren’t perfectly right “they were defensible and at least spurred the thinking in the right direction, which for an outside-in ten-minute scan is about as good as you can get.”
5. The same seven, weighted differently by industry. Medical technology versus automotive components: brand strength matters more in medtech because “if you have a reputation for poor quality, you just don’t survive… no physician is going to want to use your products”; innovation and IP carries a higher premium there too, where “a difference of a few percentage points or few outcomes does actually move the needle,” while in automotive components “you have to be really, really good to truly make that a differentiated point that your customer is going to pay for.”
6. The asymmetry finding — the episode’s most actionable result. Advantages cluster and compound: a company strong across four areas of differentiation “really outperformed quite a bit. Only one area didn’t.” But deficits are punished far harder than strengths are rewarded. Companies with no deficit areas did “all right… they didn’t drastically outperform, but they did okay.” Whereas “if they have a deficit in one of the seven areas, economic profit goes from slightly positive to massively negative. And if they have two more, it almost doubles again.” Banholzer’s rule follows: “you have to have an advantage and you have to be good enough on everything else. If you have the best intellectual property or product in the world, but your sales team really underperforms and they can’t get the product in front of customers, that doesn’t matter.”
7. The corollary about human nature, and the chemical-company case. “Everyone likes to talk about what they’re good about and not really focus on their weaknesses, and you really need that cold-eyed view.” His worked example: a chemical company that believed it was best-in-class at a specific type of polymer innovation “were over-delivering on that one piece. What they really needed to do was bring the rest up to parity and fix a deficit on their route to market.” Explicitly against the personal-development analogy: “at a personal level [we] talk about strength-based growth… we’re finding at an enterprise level, you do need to actually address your weaknesses because it will hamper your strengths.”
8. Three decision axes: efficiency, durability, extensibility. Efficiency — capital allocation with two failure modes on either side: over-engineering past the point customers pay for (“you can go too far and hit a diminishing-returns area”), and neglecting something until “it actually starts to destroy the value you’ve created by building out some other strength.” Durability — “how long is it likely to last? Are there trends that are about to topple you? Have you taken your foot off the gas and your competitors are catching up?” Extensibility — “wherever your advantage is right now, can it be taken somewhere else?” With the constraint stated plainly: “a cement company can’t usually win successfully moving into software.” The pay-off is choosing adjacencies “where what you’re good at is valued… rather than entering a market that might not value it, and it’s sort of commoditising.”
9. Cadence: monthly to quarterly, not annual. “The cadence for the companies that do this really well and grow very well is much faster than annually. It’s usually quarterly or monthly” — and not a clean-sheet exercise each time but continuous tracking “so that as things start to spike, they see them with enough lead time that they can do something about it.” Banholzer’s framing of what makes it possible now: this can be done “in a way that is much easier and practicable than it was even five years ago.”
10. Why advantage is eroding, and the early-warning problem. LaBerge: “economic profit is a bit of a lagging indicator… we’ve seen [organisations] where they don’t see that their competitive gap is closing until it already topples.” Her reference point is the Hemingway line — “how did you go bankrupt? Very slowly, and then very fast.” Causes named: “the overall porosity of industry barriers, really mixing up who you’re competing with and what your customers are expecting,” plus AI-specific disruption. Two failure signatures worth recording: a strength turning into a weakness (a company with mounting operational issues that “had been a strength and now it was not”), and an advantage held but not deployed — “financial strength where you have these enormous cash assets, but you’re not actually using that for anything… you’re not deploying it. And that can be true for any of the seven.”
11. How AI changes it — speed, not depth. “I wouldn’t say it’s broader or deeper necessarily than some of these other changing innovations that have happened in the past, but you can’t argue that it’s not faster.” It hits each of the seven differently: brand and reputation via trust; scale via how fast ecosystems can be accessed and what they are made of; operations most visibly in pharmaceuticals, “where for so long it was laboratory-based experimentation which was core to their R&D function, and now more and more of that is going to innovation and R&D through simulations and AI — how is that disrupting and changing the core competencies required there?”
12. Two extended worked examples. Disruption: continuous glucose monitoring displacing fingerstick strips. The incumbents had dominant share, established payer reimbursement pathways and a consumables business model; CGM arrived with “data ecosystems, a very different type of device that was embedded, and a different reimbursement model.” Her point is about which signal mattered: “there were lots of new technologies always coming online. Not everything wins, not everything scales. The key tipping points started to be when you got some of the regulatory changes passed and the reimbursement permissions changed — and that preceded the massive tip by a few years. Enough time that you can try to pivot or decide to exit.” Extension: an automotive-and-assembly supplier making tiny, low-vibration, quiet motors for combustion engines in a flat market. Conventional adjacency analysis offered little; a competitive-advantage lens surfaced heart pumps and medical devices — a perfect capability fit at a much higher growth rate. But they went in knowing what they lacked: no reputation in medical, no equivalent OEM relationships, and “a totally different metabolic rate” of innovation than the niche where “they were kind of king of the hill.”
13. The AI-era prescription — encode the advantage as an ontology. Banholzer’s answer to how do you track this holistically turns into the episode’s most forward-looking passage: companies are “really think[ing] about this idea of a semantic layer, or what us and others have called an ontology, to think through how do you really codify what makes your business unique and special.” The content is decision structure, not documents: “if your advantage is how you manage your supply chain, there are real decisions you probably do make in terms of which plant gets allocated which resource, how you do S&OP planning, how you prioritise which customers… how do you map the connections of how decisions flow through the organisation — what people take what decision to do what, what asset moves what resource from one place to another.” He calls the artifact “almost a digital twin to your operating model,” and makes three claims for it: it is itself hard to replicate; it “creates the railroad tracks that underpin how different AI agents can navigate… to prevent hallucination” and supplies guardrails for analytics; and it addresses the talent cliff — “a lot of great people who are retiring soon; how do you encode their knowledge?” His closing note ties it back to extensibility: “if you have this, it’s easy to apply that agent to a new market, a new industry.”
14. The five questions to close on. (a) Do you actually know what matters to your customers? — “companies think it’s something and either their customers don’t agree that they’re good at it, or they’re good at it but the customer doesn’t care.” (b) Do you know how good is good enough? — “you don’t need to win on absolutely everything… you just have to understand where is that zone where you don’t need to over-invest,” while not negatively differentiating either. (c) Are you gaining or losing ground relative to competitors in the plays you’re currently in? (d) What trends should you be monitoring, and continuously or twice a year? — “do you understand what that rate of change in your business is, and how to get at that data?” (e) When you make a move, are you using competitive advantage to inform it?

What was actually ingested

The full auto-generated (ASR) English caption track (461 segments, consistent with duration: 48:12 / length_seconds: 2892). All 12 chapter markers present. Speaker turns are unlabelled; the three-voice structure (host plus two guests) is mostly unambiguous because Brown addresses each guest by name at handover, but a small number of short interjections are ambiguous between Banholzer and LaBerge and are attributed above only where the turn structure is clear. The show’s standard open, sign-off and subscription trail are excluded from the substantive summary. The underlying McKinsey article on the erosion of competitive advantage is referenced but not ingested — every statistic here is the guests’ verbal characterisation of it.

Dynamic-capabilities tagging

• digital-sensing/digital-scouting — the core methodological claim is that the seven sources of advantage are now detectable outside-in, across the top 5,000 global companies, in what LaBerge calls “an outside-in ten-minute scan.” The prescribed practice is explicitly a scanning discipline: batch-mode scans for new patents, competitive offerings and rival capex shifts, plus “always-on” tracking of spikes in startup activity and regulatory news, with triggers cascaded into the business.
• digital-sensing/digital-scenario-planning — the durability axis is scenario work in operational form: triggers “hit how your strategic scenarios look” and “hit indicators for should we revisit our resourcing decisions.” The CGM example is the method’s worked case, where regulatory approval and reimbursement changes were the signals that “preceded the massive tip by a few years.”
• digital-seizing/balancing-digital-portfolios — efficiency is stated as a capital-allocation problem with a two-sided failure mode (over-engineering past what customers pay for; neglecting a capability until it destroys value built elsewhere), and the deficit asymmetry finding turns it into a concrete allocation rule: fix deficits before pushing strengths further, because a single deficit moves economic profit “from slightly positive to massively negative.”
• strategic-renewal/business-model — extensibility is a business-model-renewal test rather than a market-attractiveness one (“a cement company can’t usually win successfully moving into software”), with the tiny-motors-into-medical-devices case as a worked example of choosing an adjacency where the existing advantage is valued rather than merely applicable.
• contextual/external-triggers — the erosion thesis is an external-conditions argument: an 11% rise in shuffle rate across 60%+ of industries, “the overall porosity of industry barriers really mixing up who you’re competing with,” and AI as an accelerant that “hits each of the seven elements in different ways.”

Linked entities and concepts

• McKinsey & Company — publisher; Inside the Strategy Room, a sixth McKinsey channel in the wiki. Updated in this ingest.
• MotherDuck — the same ontology-and-semantic-layer prescription from the data-platform side; see this source’s relationships:.
• Catlin — the same firm’s durable-advantage argument, with this episode supplying the measurement layer.
• McGrath — transient advantage requiring continuous re-examination, here quantified.
• Frey — decaying incumbency advantage read at a different level.
• strategy — the definition with its three qualifiers, the seven sources, the efficiency/durability/extensibility frame, and the five closing questions.
• strategic-foresight — economic profit as a lagging indicator, the batch-versus-always-on signal taxonomy, and the CGM case’s finding that the actionable tipping point was regulatory and reimbursement change rather than the technology’s arrival.
• knowledge-graphs — the ontology/semantic-layer prescription, framed as encoding decision flow rather than documents, and as “a digital twin to your operating model.”
• enterprise-ai-adoption — the ontology as guardrail infrastructure for agents (“the railroad tracks… to prevent hallucination”) and as an answer to the retiring-expert talent cliff.
• dynamic-capabilities — the five cells tagged above.
• theory-based-view — the insistence that a competitive-advantage claim must be testable against economic profit rather than being “yet one more framework.”

Dangling (single-source mention, deferred per author-entity promotion): Matt Banholzer, Laura LaBerge, Sean Brown.

Source quality note

Auto-generated transcript; proper nouns corrected at acquire time (McKinsey, Sean Brown, Laura LaBerge, Matt Banholzer, laggards, dot-com, tranche, parity, moats, porosity, S&OP, fingerstick — see the raw file’s notes:).

This is McKinsey publishing its own research on its own channel, and the research underpins a service line (strategic growth and innovation). Banholzer pre-empts the obvious objection — “what we don’t want to do is just make this another exercise that creates yet one more framework… make sure this is truly data-driven and linked to what’s out there” — which is a better posture than the format usually produces, but the wiki cannot check it: the article is not ingested, so the 5,000-company sample, the 180+ metals-and-mining set, the 11% shuffle-rate increase, the 10%-alignment and two-thirds-track-only-at-BU-level figures, and above all the deficit-asymmetry result are all verbal characterisations of unexamined work.

Two specific caveats. The seven sources are given as a list in speech and never enumerated cleanly in one place; the enumeration above is reconstructed from scattered mentions across the episode and should be checked against the article before being quoted as canonical. And the outside-in detection claim — that these markers are legible from public data via AI-assisted analysis — is the load-bearing methodological premise for the entire product, validated in the episode only by “for the companies that we did know or had experts on… they were defensible,” which is a low bar stated honestly. Client cases (the chemical company, the motor supplier) are anonymised and unquantified.