Strategy & competition
Core Competence Analysis
A test of what a corporation is genuinely good at: the collective learning and coordination skills that pass three tests by opening access to a wide variety of markets, contributing significantly to perceived customer benefit, and resisting imitation.
Also known as Core competencies analysis, The three tests of core competence. First set out by C. K. Prahalad and Gary Hamel in 1990; the primary source is cited in full below.
- Format
- Checklist / audit
- Level
- Corporate · Business unit
- Best for
- Position against competitors · Allocate resources · Evaluate options
- Decision stage
- Diagnose · Decide
- Difficulty
- Intermediate
- Time to apply
- Half a day for a first screening with honest evidence; validating imitability claims takes longer.
Plate · The model
The components
Provides access to a wide variety of markets
The first test: a core competence opens doors beyond its original product line. If the capability cannot plausibly seed offerings in several markets, it may be a strength, and it is not core in Prahalad and Hamel's sense.
Signals of strength
The capability already shows up in more than one product or business · Credible adjacent markets would pay for what it produces · The skill transfers without the original product attached
Makes a significant contribution to perceived customer benefit
The second test: the competence must drive benefits customers actually perceive in the end product. Capabilities that are essential yet invisible to customers, however well executed, fail this test.
Signals of strength
Customers cite the resulting benefit unprompted · Removing the capability would visibly degrade the offer · The benefit shows up in willingness to pay or in loyalty, with evidence
Is difficult for competitors to imitate
The third test: rivals should find the competence hard to replicate. Difficulty is greatest when the competence is a complex harmonisation of technologies and production skills built through accumulated learning, with no single blueprint to steal or buy.
Signals of strength
Built through years of accumulated learning rather than purchasable inputs · Woven through many teams and processes rather than held by one · Competitors have tried to copy it and shipped inferior versions
When it earns its keep
- You are deciding what to keep in-house and what to outsource, and need to know which capabilities are the crown jewels before signing anything away.
- Diversification is on the table and you want to know which adjacent markets your existing capabilities give you a right to win, rather than merely a desire to enter.
- Business units are hoarding people and budgets, and you suspect the corporate whole is failing to invest in competences that outlast any single product.
- The organisation's story about what it is best at has never been tested against evidence, and strategy is being built on the untested version.
And when it doesn't
- You are a young single-product company. Competences are accumulated collective learning; a startup is still acquiring them, and discovery tools fit the moment better.
- You need the market-back view. The analysis looks inward at capability; pair it with Five Forces or PESTEL before concluding that a market wants what you are good at.
- The exercise is being used to bless the status quo. If every current activity ends up declared core, the tests have been inverted into a loyalty oath.
- You need resource-level granularity. VRIO tests individual resources and capabilities; core competence works at the level of integrated bundles of skill and technology.
How to run it
Before starting, gather the inputs the analysis depends on:
- An inventory of candidate competences framed as collective learning and coordination skills, rather than products, assets or single technologies.
- Customer evidence showing which benefits each candidate actually drives, from research rather than internal belief.
- Competitor intelligence on who else can do this, how well, and how quickly they could learn it.
- A map of which end products and business units draw on each candidate competence.
- 1
Inventory candidate competences
List candidates at the level Prahalad and Hamel define: the collective learning in the organisation, especially the capacity to coordinate diverse production skills and integrate multiple streams of technology. They judged that few companies could build world leadership in more than five or six; a list of fifteen is a skills audit, and the framing has gone wrong.
- 2
Apply the market-access test
Ask whether the competence provides potential access to a wide variety of markets. Their example is Honda's engine competence carrying it from motorcycles into cars, generators and lawnmowers. A capability locked inside one product line may be valuable and still fail this test.
- 3
Apply the customer-benefit test
Ask whether the competence makes a significant contribution to the perceived customer benefits of the end product. The judge is the customer, which is why this test needs external evidence rather than internal conviction.
- 4
Apply the imitability test
Ask whether the competence is difficult for competitors to imitate. Prahalad and Hamel argue imitation is hardest when the competence is a complex harmonisation of individual technologies and production skills: a rival can buy the same equipment and still fail to copy the coordination.
- 5
Trace competences to core products
Identify the core products that physically embody each competence and check whether the corporation is investing in them, or only in end products. Underinvestment at this layer is invisible in any single business unit's accounts and fatal in aggregate.
- 6
Set the stewardship agenda
Decide which competences to build, who carries them, and how business units will share them. The article's argument is that competence carriers are corporate assets, and that a strategic architecture should govern their deployment across unit boundaries.
Reading the result
A short list of genuine core competences, often between zero and three when the tests are honestly applied, together with the core products that embody them and an agenda for investing in, protecting and reusing them across markets.
- Surviving all three tests is rare by design. Prahalad and Hamel doubted any company could lead in more than five or six competences; a long list means the bar was set at 'things we do' rather than 'things that pass the tests'.
- A candidate that fails one test is information. Failing market access leaves a niche capability to exploit where it stands; failing imitability leaves a cost of entry, with a countdown until rivals catch up.
- Read the result at corporate level. The article's target was the tyranny of the strategic business unit, which optimises end products while starving the shared competences beneath them.
A worked example
A drone surveying company tests what it is actually selling
A thirty-person UK drone surveying firm serves construction, infrastructure and quarrying clients with topographic surveys, volumetric measurement and structural inspection. Growth has stalled, and the directors disagree about expanding into agriculture, wind-farm inspection, or white-labelling flight services for other surveyors. Before choosing, they run their capabilities through the three tests.
- Inventory candidate competences
- Three candidates survive the first framing: certified flight operations at scale, rapid sensor and payload integration, and the photogrammetry processing pipeline that turns raw imagery into survey-grade deliverables with documented accuracy. A fourth nomination, 'our client relationships', is set aside as an outcome of competence rather than a competence.
- Provides access to a wide variety of markets
- Flight operations fails: every operator holds comparable CAA authorisations, and the capability opens no doors rivals cannot also walk through. The processing pipeline passes strongly: accuracy-assured measurement from aerial imagery is valued in agriculture, mining, insurance loss assessment and renewables inspection. Sensor integration passes moderately, largely as an enabler of the pipeline.
- Makes a significant contribution to perceived customer benefit
- Clients do not buy flights; they buy defensible measurements their engineers can sign off. Tender feedback and renewals repeatedly cite the documented accuracy and the audit trail behind each deliverable. Flight operations is invisible when it works, contributing safety and legality that clients assume rather than perceive.
- Is difficult for competitors to imitate
- Drones, cameras and off-the-shelf photogrammetry software are available to anyone, and operational authorisations take effort without forming a moat. What resists imitation is seven years of accumulated processing know-how: error budgets by terrain type, calibration practice, and QA routines refined across two thousand jobs. Rivals with identical hardware deliver visibly noisier data.
The read. Only the processing and accuracy-assurance capability passes all three tests. The reading is uncomfortable and useful: the firm has been marketing itself as a drone company while its core competence is measurement assurance, with drones as one capture method among several. Expansion should follow the competence into wind-farm inspection and mining volumetrics, where accuracy commands a premium. The white-label flight idea is dropped, since it scales the imitable part of the business while lending competitors access to none of the defensible part.
Pitfalls
- Listing products, assets or single technologies as competences. The definition is collective learning and coordination; a patent or a machine may sit inside a competence, and neither is one.
- Letting every function nominate its own work as core. The three tests exist to cull; an uncut list is a morale document rather than a strategy.
- Confusing 'important' with 'core'. Payroll must run flawlessly and passes none of the tests; the framework separates what must be done well from what wins markets.
- Ignoring the flip side of competence. The same embedded capability can harden into rigidity, so test whether yours remains an asset in the market you are entering.
- Treating the analysis as one-off labelling. The article's demand was ongoing stewardship: invest in competences, protect their carriers, and lend them across business units.
What the critics say
Coyne, Hall and Clifford argued that for most companies core competence proves a mirage: executives applying the tests either find nothing that genuinely passes or settle for bland self-descriptions, and the concept offers little guidance on how to build a competence you lack.
Coyne, K. P., Hall, S. J. D. and Clifford, P. G. (1997) 'Is Your Core Competence a Mirage?', The McKinsey Quarterly, No. 1, pp. 40–54.
Leonard-Barton showed that the deeply embedded capabilities that pass the tests can simultaneously operate as core rigidities: the skills, systems and values that made a firm strong actively inhibit the new product development that changing markets demand.
Leonard-Barton, D. (1992) 'Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development', Strategic Management Journal, 13(S1), pp. 111–125.
The reasoning risks circularity in practice. Competences are most often identified retrospectively in already-successful firms, which makes the logic hard to falsify and easy to flatter; prospective identification, before market success confirms the judgement, is where the framework is weakest.
Sources and further reading
- Prahalad, C. K. and Hamel, G. (1990) 'The Core Competence of the Corporation', Harvard Business Review, 68(3), May–June 1990, pp. 79–91. ↗
- Hamel, G. and Prahalad, C. K. (1994) Competing for the Future. Boston, MA: Harvard Business School Press.
- Coyne, K. P., Hall, S. J. D. and Clifford, P. G. (1997) 'Is Your Core Competence a Mirage?', The McKinsey Quarterly, No. 1, pp. 40–54.
- Leonard-Barton, D. (1992) 'Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development', Strategic Management Journal, 13(S1), pp. 111–125.