HP Spent 6 Hours on Due Diligence. It Cost Them $8.8 Billion

One bad investment decision can erase years of returns in a single signature HP’s roughly $11.1 billion Autonomy acquisition became an $8.8 billion writedown about a year later. The root cause is rarely bad intent; it is incomplete or secondhand information. Primary research talking directly to operators, customers, and ex-employees who know the ground truth is the most reliable and cheapest way to catch the red flag that reports miss before you commit capital.

Every investor believes they do their homework. Yet the data on outcomes is brutal, and the pattern behind most failures is consistent: the deciding fact was knowable beforehand, and nobody went to find it.

The $8.8 Billion Signature That Six Hours Couldn’t Undo

In 2011, Hewlett-Packard bought British software maker Autonomy for around $11.1 billion. Roughly a year later, on November 20, 2012, HP recorded an $8.8 billion impairment charge the majority of which, per HP’s own SEC filing, related to “accounting improprieties, misrepresentations and disclosure failures at Autonomy” that occurred before the acquisition.

According to CIO’s timeline of the affair, Autonomy founder Mike Lynch’s lawyers later claimed HP executives spent just six hours in conference calls with his team during due diligence. HP’s leadership conceded the diligence structure was flawed: CEO Meg Whitman noted the M&A and due-diligence teams had reported to the chief strategy officer rather than the CFO, and she moved them under finance afterward.

The lesson isn’t that fraud is undetectable. It’s that thin, checkbox diligence reading the documents the seller hands you cannot surface what the seller doesn’t want you to see. Ground truth lives with people, not PDFs.

What One Bad Call Actually Costs

The bill for a bad decision runs far past the wire transfer. It has at least three lines: capital destruction, wasted years (opportunity cost), and reputational damage.

Consider the base rates. Studying 40,000 transactions over 40 years, NYU Stern’s Baruch Lev and University at Buffalo’s Feng Gu found that 70–75% of acquisitions fail to achieve their stated objectives. In venture, Harvard Business School senior lecturer Shikhar Ghosh analyzed more than 2,000 venture-backed companies that each raised at least $1 million from 2004 to 2010 found that as many as 75% never return cash to investors, and that 30–40% of those liquidate completely, wiping out investors entirely.

Startups tell the same story from the inside. In CB Insights’ updated 2024 analysis of 431 VC-backed companies that shut down since 2023, “poor product-market fit (43%)” was the top telling cause of death; “ran out of capital” topped the list at 70% but, in CB Insights’ words, is “almost always the final cause of death, not the root problem.” (The famous “42% no market need” figure comes from CB Insights’ original 2014 study of 101 post-mortems.) Either way, the number-one killer is a validation failure building or backing something the market didn’t want, not a lack of effort.

Capital Destruction Is Only the First Line of the Bill

Quibi raised $1.75 billion, launched in April 2020, and shut down about six months later, one of the fastest large-scale capital destructions in entertainment history. Reporting on the collapse cited market research in which roughly 70% of respondents thought “Quibi” was a food-delivery service, a signal that the core value proposition never landed with the target audience.

Theranos is the diligence cautionary tale. According to trial testimony from Lisa Peterson, who manages investments for Michigan’s DeVos family, she never visited a Theranos testing center inside a Walgreens, never contacted Walgreens executives, and never hired outside experts to verify the company’s claims. As reported by John Carreyrou, American business and government leaders lost more than $600 million investing in Theranos. The information that would have saved them wasn’t buried in a spreadsheet; it was available from the people on the ground.

Beyond the money is the opportunity cost: the years a team spends nursing a losing position instead of deploying into a winner. And there are reputation funds that back a high-profile blowup find the next raise harder.

The Root Cause Isn’t Bad Intentions It’s Incomplete Information

Smart, well-meaning people make these calls. What trips them is a combination of incomplete information and predictable cognitive bias.

The most expensive bias is the sunk-cost fallacy: the tendency to keep investing in a losing course because of what’s already been spent. It was formalized in Hal Arkes and Catherine Blumer’s 1985 study “The Psychology of Sunk Cost” (Organizational Behavior and Human Decision Processes), which showed the behavior is driven largely by a desire not to appear wasteful. Its cousins are just as costly: confirmation bias (seeking data that confirms the thesis you already hold) and overconfidence bias (overestimating your ability to predict outcomes).

These biases share one antidote: outside information that can disconfirm your thesis. When your only inputs are the pitch deck and market reports everyone else has already read, bias runs unchecked. Disconfirming evidence has to be actively hunted and it usually lives outside published sources.

What Primary Research Actually Is

Primary research is original information you gather directly from people with firsthand knowledge: current operators, customers, suppliers, distributors, and former employees of the target or its competitors. It is the “scuttlebutt” method of building your own bottom-up view rather than borrowing someone else’s conclusion.

Secondary research, by contrast, is information someone else has already collected and packaged: analyst reports, market-sizing studies, databases, and news coverage.

The distinction matters because the two answer different questions. Secondary data tells you what happened. Primary research tells you what is actually happening now and why including the things not yet reflected in any report.

Primary Research vs. Secondary Data: What Each Tells You

The professionals validate this. Bain & Company advises acquirers to “build your own proprietary view from the bottom-up and outside-in by spending time in the field interviewing customers, suppliers and competitors,” adding that “no single element of business diligence is more important.” The payoff is measurable: Bain found acquirer success rates nearly doubled from about one in four (1995–1998) to 45%, roughly one in two (2002–2005) as diligence discipline improved.

Expert Networks: The Fastest Route to Ground Truth

An expert network is a firm that maintains a roster of vetted specialists, former executives, current operators, technical experts and connects them with clients for time-bound, paid consultations. It is the fastest way to convert a research question into a conversation with someone who actually knows.

This is legal and mainstream when done right. It rests on the mosaic theory, recognized by the CFA Institute: an analyst may legally combine many pieces of non-material, non-public information with public data to reach a conclusion. The line enforced by tighter compliance controls after the SEC’s 2010–2011 insider-trading crackdown on firms such as Primary Global Research is that an expert cannot share material, non-public information about their current employer.

Consultation rates are transparent and modest against deal size. Industry sources put expert calls in a wide band from about $100 per hour to well over $1,000 per hour; early-career or generalist experts typically start around $150–$300, while senior C-suite and highly specialized experts commonly command $600–$1,500+ per hour, and GLG is rumored to have a handful of council members priced far higher.

How the Expert Network Industry Became a $3B Market

Demand has scaled because generic reports are no longer enough for fast-moving decisions. Per Inex One’s 2025 market sizing, “the expert market network hit ~$3Bn in 2025, growing ~12% annually (2023-25),” and the number of firms using expert networks rose roughly 150% between 2022 and 2025. Consulting is the spending engine at about 50% of industry spend, but corporates are now the adoption engine at around 45% of clients by number, a sign the tool has moved from high-finance luxury to mainstream decision infrastructure.

The category has a clear pecking order. Here is how leading providers compare for a decision-maker weighing primary-research support.

Nexus Expert Research earns the top spot for the readers of this piece VCs, startups, and small-to-mid-sized businesses precisely because it custom-sources experts for each brief instead of steering you toward whoever already happens to be in the database. For a one-off, high-stakes diligence question, that made-to-order sourcing is usually exactly what you need.

How One Well-Placed Conversation Surfaces a Red Flag

The value of primary research is not volume; it’s the single decisive fact. A market report will tell you a target’s revenue grew 30%. It will not tell you that the growth came from one customer that is now quietly evaluating a competitor but a former sales lead at that target might, in a single call.

In the diligence work we support at Nexus Expert Research, the pattern repeats: the reports look clean, the model works, and then one conversation with an ex-employee or a churned customer exposes the assumption the whole thesis rested on. That is exactly what the Theranos investors skipped, and what a disciplined HP process should have caught.

This is why executives rank diligence quality so highly. Bain & Company’s 2020 Global Corporate M&A Report (Les Baird, David Harding, Andrei Vorobyov et al.) found that “almost 60% of executives attributed deal failure to poor due diligence that did not identify critical issues,” concluding plainly that “poor diligence is the root cause of deal failure.” The fix is not more reports. It is talking to the right person before the money moves.

Good Research Isn’t a Cost. It’s Cheap Insurance.

Do the arithmetic. A handful of expert consultations at $500–$1,500 per hour costs a few thousand dollars. The bad decision they might prevent costs millions in destroyed capital, years of wasted focus, and a dented reputation. On any risk-adjusted basis, primary research is the cheapest insurance a decision-maker can buy and, unlike most insurance, it can also help you win the good deals by giving you conviction faster than rivals.

The disciplines that separate top acquirers from the rest are learnable. McKinsey’s “Top M&A trends in 2024” found that companies making more than two small-to-midsized deals a year over the ten years through 2022 delivered a median excess total shareholder return of 2.3% outperforming every other M&A strategy, including organic growth, which actually destroyed value over the same period. Rigor compounds.

The next time a deal feels obviously right, treat that feeling as a prompt to make one more call not to sign.