When Procter & Gamble publishes its sourcing principles, it states plainly that it "prefers ongoing relationships with incumbent suppliers" and that new suppliers must offer "meaningfully better total value than the incumbent" to win business. This is not a secret. It is the written policy of one of the world's most sophisticated procurement organizations. The bar for challengers is higher by design.
Most organizations do not write this policy down. But it operates anyway — through evaluation criteria that reward familiarity, through relationships that blur the line between trust and scoring, and through processes that give the incumbent information advantages no challenger can match.
What incumbency bias actually is
Incumbency bias is the systematic preference for a current supplier beyond what their price, performance, or total value objectively justifies. It is not corruption. It is not lazy procurement. It is the accumulation of small structural advantages — information access, relationship depth, institutional familiarity — that tilt competitive processes before a single bid is evaluated.
In auction theory, this is well understood. Research on procurement auctions documents "incumbency advantages that prevent entrants from submitting competitive bids," according to the International Journal of Industrial Organization. The incumbent knows the specifications better. The incumbent understands the buyer's unwritten preferences. The incumbent has a track record — and track records are weighted even when the question is about future performance, not past.
How the advantage compounds across evaluation stages
The bias operates through three distinct mechanisms, each reinforcing the next. First: information asymmetry. Incumbents understand the buyer's actual consumption patterns, not the estimated volumes in the RFP. They know which service levels matter and which are never enforced. Challengers bid against a document; incumbents bid against reality.
Second: specification capture. When RFPs are written, the specifications often describe what the incumbent already delivers. Performance criteria, technical requirements, service models — they are extracted from the existing contract and republished as neutral requirements. A challenger bidding against a spec built on the incumbent's operating model starts at a structural disadvantage.
Third: evaluation weight distribution. Most scoring models assign points for "relevant experience," "demonstrated capability," and "relationship management." These criteria are defensible individually. Collectively, they give 20-30% of the total score to factors where the incumbent has an inherent, non-replicable advantage. A challenger can beat the incumbent on price and still lose on the weighted score.
Where the bias does the most damage
The financial harm is concentrated in categories where switching costs are low but trust costs are perceived as high. Professional services, facilities management, logistics, and IT managed services are notorious. In these categories, the fear of disruption — often exaggerated — keeps incumbents in place through multiple contract cycles without meaningful price or performance challenges.
Specialist federal contracting analysts report that incumbents win roughly 80% of competitive recompete actions in US federal procurement, based on FPDS-NG data and GAO protest pattern analysis. This is not a myth. It is a structural feature of how government acquisition works — and the private sector is not immune. Huthwaite International's global study across Europe, the US, and Australia found incumbent retention at 54.2%, lower than the federal rate but still a clear majority advantage.
Why evaluation processes reinforce the bias without detection
Most procurement teams believe their evaluation processes are objective because they use scoring matrices. But a scoring matrix that assigns 15 points to "understanding of our business" and 10 points to "cultural fit" is not objective — it is a weighted preference for incumbency disguised as rigor.
Relationship-weighted scoring where "familiarity" and "past performance" consume 20-30% of total evaluation points. Evaluators know which bid is the incumbent. No win-rate tracking. No blind review rounds.
Criteria separated from relationship: past performance scored on documented data, not impression. Blind evaluation rounds where supplier identity is hidden. Win rates tracked and reviewed quarterly.
Without measuring the outcome, the bias remains invisible. Organizations that do not track incumbent win rates cannot know whether their process is competitive. And most do not track it. The bias protects itself by never generating a number that would expose it.
What correct execution looks like
Organizations that reduce incumbency bias do three things differently. First: they measure it. Tracking incumbent win rates by category — and reviewing the data quarterly — makes the bias visible. A CPO who sees 85% incumbent retention in IT services but 52% in direct materials knows where the evaluation process is broken.
Second: they separate relationship from scoring. Past performance is scored on documented data — SLA compliance, delivery metrics, quality audit results — not on evaluator impression. "Understanding of our business" is removed from the scoring matrix entirely and replaced with a structured knowledge transfer requirement that any qualified supplier can meet.
Third: they introduce blind evaluation rounds. At least one scoring stage is conducted without supplier identification. Evaluators receive anonymized technical responses. The incumbent's name is not a variable. This single change eliminates the familiarity premium and forces scoring against the stated criteria, not the known entity.
- Track incumbent win rates by category and review quarterly — the number itself creates pressure to improve
- Audit evaluation criteria: remove "understanding of our business," "cultural fit," and similar familiarity-weighted factors
- Score past performance on documented data (SLA reports, quality metrics), not evaluator impression
- Run at least one blind evaluation round per sourcing event where supplier identity is hidden
- Require a written justification when the incumbent scores above 80% on non-price criteria
- Benchmark challenger disqualification rates against incumbent — if challengers are eliminated at 3x the rate, the process is filtering for familiarity
FAQ
Is incumbency bias always bad?
No. There are legitimate reasons to prefer incumbents: proven performance, lower transition risk, institutional knowledge. The problem is not the preference — it is the lack of measurement. If an incumbent retains a contract because they deliver superior value, that outcome should be visible in the data. If they retain it because no one tracked the win rate, that is the bias operating undetected.
How much does incumbency bias cost?
Organizations that implement structured bias-reduction practices — tracking win rates, blind evaluation rounds, documented past-performance scoring — report 8-15% savings on recompete categories. The cost is the spread between what the incumbent charges and what a competitive process would produce, compounded across multiple contract cycles.