Why 86% of ODI-Guided Products Succeed (And What That Means for You)

A Statistic You Should Question — and Then Take Seriously

86%.

That is the product success rate that Tony Ulwick and die ODI-Praxis claim for products developed using the Outcome-Driven Innovation methodology. Compared to the commonly cited industry average of 17% — or, depending on the study, somewhere between 10% and 30% — the 86% figure is so dramatically different that it invites skepticism.

You should be skeptical. No claim this striking should be accepted uncritically, especially one that comes from the inventor of the methodology. Let me show you why, after examining the evidence carefully, I am convinced the directional claim is correct — and why the specific number, while not independently audited, is plausible given the mechanism behind it.

I have been running ODI engagements for over 20 years. I have seen products succeed that should not have, and I have watched products with excellent technical specifications fail commercially. The pattern is consistent enough that the causal mechanism behind the 86% claim — not the specific number, but the mechanism — is not in serious doubt. The question worth examining is why it works, not whether it works.

The Industry Baseline: Why Product Failure Is Normal

Before examining the 86% claim, the baseline needs to be established. What is the typical product success rate that ODI is being compared against?

The Evidence Base

The research on product failure rates spans decades and multiple methodologies. Some key data points:

• Clayton Christensen (2011): Cited a figure of roughly 85% of new products failing to meet their targets — a figure he often used in lectures and which entered broad circulation.
• Nielsen (2014): Reported that 85% of new consumer goods products fail within the first two years.
• Castellion and Markham (2013): Reviewed the literature on product failure rates in the Journal of Product Innovation Management and concluded that the commonly cited figure of “90% failure rate” was not empirically well-supported, but that failure rates in the range of 40-60% for new products were well-documented even by conservative definitions.
• Product Development and Management Association (PDMA) studies: Repeatedly found success rates in the 50-65% range for products that make it to market — but with a significant caveat: many studies count only products that reached market launch, excluding the 50-60% of projects that are cancelled before launch.

When you include cancelled projects in the denominator — which is the honest way to measure — the overall probability that a product concept generates commercial success is closer to 15-30% depending on the industry.

For industrial and B2B products, the failure rates are somewhat lower than consumer goods (engineers are better at building things that work) but the commercial failure rates — products that launch but do not meet revenue targets — are similar. Several studies of B2B product launches find commercial success rates of 25-35%.

The baseline, then, is somewhere in the range of 17-33% depending on definition and industry. The ODI claim of 86% is compared against this baseline.

The 86% Claim: What It Means and Where It Comes From

Ulwick’s Evidence Base

Ulwick’s 86% figure comes from the published ODI track record. Over 30+ years of practice, across more than 400 ODI projects, Practitioners have tracked the commercial performance of products developed using the methodology. The 86% represents the fraction of those products that met or exceeded their commercial targets within the first three years after launch.

Several important caveats apply:

It is not independently audited. Practitioners track their own client results; this is not a third-party retrospective study. The potential for selection bias — clients who succeeded are more likely to maintain the relationship and share data; clients who failed may attribute failure to execution rather than methodology — cannot be dismissed.

“Success” is defined by client-reported commercial performance. There is no standardized definition across the portfolio. A product that met a modest revenue target counts the same as one that exceeded a stretch target.

The denominator matters. If the published ODI track record skews toward sophisticated, well-resourced organizations with strong execution capabilities, the 86% may reflect a high baseline of execution quality rather than the methodology alone.

These caveats acknowledged, the claim survives scrutiny for a specific reason: the mechanism by which ODI should increase success rates is theoretically sound, empirically grounded, and observable in practice.

I do not use the 86% figure as a sales pitch. I use it as a prompt. It prompts the question: if an approach that addresses the root cause of product failure has a documented track record of dramatically better outcomes, what is the opportunity cost of not using it? That is the conversation worth having.

Why ODI Should Increase Success Rates: The Mechanism

The reason the 86% claim is plausible — despite its limitations as a statistic — is that it is grounded in a clearly identified mechanism. Product failure has known, documented causes. ODI directly addresses those causes. If the methodology is executed correctly, the mechanism should produce better outcomes.

Mechanism 1: Eliminating the Input Problem

Research on why products fail consistently points to a fundamental problem with how product requirements are gathered. Products fail commercially because they address the wrong needs — not because they are built poorly. The most common failure modes are:

• Building what customers asked for, not what they needed. Customers request solutions based on their awareness of what is possible; they cannot request solutions to problems they have not framed.
• Building what the loudest customer asked for. Feature development driven by vocal customers systematically over-indexes on the needs of power users and misses the needs of the broader market.
• Building what engineering wanted to build. Technology push — building features because the technology is available and exciting — produces products that are technically impressive and commercially irrelevant.

ODI addresses all three of these by replacing the input mechanism entirely. Desired outcome statements capture what customers are trying to accomplish, not what they want the product to do. The quantitative survey weights every outcome by importance across a representative population, preventing vocal-customer bias. The opportunity algorithm identifies underserved needs regardless of engineering preference.

If the reason products fail is bad inputs, and ODI produces better inputs, the products should succeed at higher rates.

Mechanism 2: Targeting Underserved Rather Than Arbitrary Needs

The opportunity algorithm is specifically designed to identify the intersection of high importance and low satisfaction — the needs that matter most and are addressed worst. A product designed to address these outcomes is, by definition, targeting the needs that the market is most eager to have solved.

This seems obvious. The counterintuitive reality is that most product development does not do this. Most product development addresses the needs that are most visible, most frequently mentioned, most requested by the most influential stakeholders — which correlates only loosely with the needs that are actually most important and most underserved.

The opportunity algorithm creates the conditions under which a product can succeed: it targets real, measurable market demand rather than organizational or anecdotal inputs.

Mechanism 3: Defining Competition Correctly

Products frequently fail because they are designed to compete in the wrong competitive space. A product that is excellent at everything a competing product does — but which does not address the needs that are driving customers to non-obvious alternatives — will underperform despite its technical superiority.

ODI’s market definition approach — defining the market around the job rather than the product category — ensures that the competitive set is correctly identified. When you know that the job of “monitor equipment performance across a distributed fleet” is being accomplished partly through your product, partly through a competitor’s product, and partly through manual spreadsheet tracking, your product design addresses the outcomes that drive customers to all three options. This is a fundamentally more complete competitive strategy.

Mechanism 4: Segment-Specific Targeting

By identifying customer segments based on patterns of underserved outcomes — not demographics — ODI ensures that product configurations are designed for segments that actually have distinct needs, not segments that look different but need similar things.

A product designed to win a specific need-based segment will win that segment decisively, because it addresses the outcomes that segment values most and that current solutions address worst. A product designed to win a demographic segment will face the same competitive challenges as a product designed without any segmentation — because demographic segments do not align with need patterns.

The Evidence Beyond die ODI-Praxis: Related Research

While the 86% figure comes from the published ODI track record, there is supporting evidence from related research:

Customer-Centric Innovation Research

Multiple academic studies (Narver and Slater, 1990; Deshpandé, Farley, and Webster, 1993; Jaworski and Kohli, 1993) have established a consistent positive relationship between market orientation — genuinely understanding and responding to customer needs rather than technology-pushing — and financial performance. Organizations that systematically measure and respond to customer needs outperform those that do not. ODI is a specific implementation of this market orientation principle.

Outcome-Based Product Design Studies

Research by Griffin and Hauser at MIT found that the completeness of customer need capture in the early stages of product development was a significant predictor of product success. Products developed with more complete understanding of customer needs — not more customer contact, but more complete need understanding — had higher success rates.

ODI’s emphasis on the completeness of outcome capture (50–150 outcomes per job, covering all stages of the Universal Job Map) directly addresses this finding. Most product development processes capture 10-20% of the outcomes that customers use to judge a product. ODI captures the full set.

The Disruptive Innovation Research

Christensen’s research on disruptive innovation, while focused on why incumbents fail rather than why new products succeed, provides indirect support for ODI’s mechanism. Disruptors succeed not because they build technically superior products but because they address needs that incumbents — focused on their most demanding customers — are systematically ignoring. ODI’s quantitative market analysis is designed to surface exactly these overlooked needs before a disruptor does.

What the 86% Actually Means for Your Organization

If you are a VP of Product or an innovation leader reading this, the 86% claim raises a practical question: what does this mean for your organization’s product development decisions?

Here is how I frame it:

The baseline cost of your current approach: If your organization launches products with a 25% commercial success rate — which is typical for industrial B2B — three out of four product investments are not generating adequate returns. The cost is not just the failed products — it is the opportunity cost of the engineering capacity, market attention, and organizational energy devoted to products that underperform.

The improvement potential: If ODI reliably produces products that succeed at three to four times the baseline rate, the expected value of ODI-guided development is dramatically higher, even accounting for the cost of the research process itself.

The calibration: You do not need to accept the 86% figure as a precise promise to conclude that ODI-informed product development should outperform intuition-based product development. The mechanism is sound. The direction of the effect is well-supported. The magnitude is uncertain — but even a 50% improvement in your success rate would transform the ROI of your product development portfolio.

How ODI-Guided Success Looks in Practice

Let me describe the pattern across the client engagements where I have seen the mechanism work:

Pattern 1: The outcome that nobody expected. In every ODI engagement I have run, the highest-opportunity outcome set includes at least one category that the product team was not addressing and did not expect to find. This is not because the team was careless — it is because the systematic, structured coverage of all job stages reveals needs at stages (typically Define, Confirm, and Conclude) that product teams consistently overlook. The products that address these unexpected opportunities succeed because they are addressing genuine unmet needs that competitors are also missing.

Pattern 2: The feature that stopped being built. ODI data consistently shows that some features currently under development — often features that engineering is excited about, or that appeared frequently in customer advisory boards — address outcomes that are already adequately served. The counterfactual is impossible to prove, but the resources redirected from these features to high-opportunity outcomes represent a significant improvement in the expected value of the development portfolio.

Pattern 3: The segment that changed everything. Opportunity segment analysis frequently reveals a segment of the market that is dramatically more underserved than the average, but that is not served by any current product. Products designed specifically for this segment — not as the universal product for everyone, but as the right product for this segment — succeed decisively in it. See how ODI case studies from client engagements illustrate this pattern across industries.

The Honest Assessment

The 86% figure should not be taken as a guarantee. No methodology eliminates execution risk. A product that addresses the right needs, built on the right technology platform, can still fail if the go-to-market strategy is wrong, if the launch timing is poor, or if the organization lacks the commercial capabilities to sell it.

What ODI eliminates — or dramatically reduces — is the specific risk of building the wrong thing. The failure mode of “we built something nobody needed” should not exist in an ODI-guided organization, because the data tells you clearly what people need before a single line of code is written or a single component is machined.

The failure modes that remain — execution risk, market timing, competitive response — are the ones that talented product teams manage well regardless of methodology. ODI does not replace that talent. It gives that talent better inputs.

The overall claim that the ODI process steps produce dramatically better commercial outcomes than alternatives is supported by the mechanism, by the related research, and by my own 20 years of practice. The 86% is the directional signal. Trust the direction.

Frequently Asked Questions