The governance vacuum after dot-com — why programme oversight started from scratch
We do not have a governance problem because governance failed. We have a governance problem because governance was never there.
The Vacuum
One year ago this week, the Nasdaq Composite Index reached its all-time high. In the twelve months since, it has lost nearly sixty per cent of its value. The companies that were going to rewrite the rules of commerce — the business-to-consumer portals, the online grocery services, the fashion e-tailers — have burned through their capital and shut their doors. The large corporates that invested in internet ventures, e-commerce platforms, and web-enabled transformation programmes are now counting the cost.
But the financial losses, significant as they are, are not the most revealing consequence of the collapse. The most revealing consequence is what the collapse exposed about how these programmes were governed — or rather, how they were not.
Across the organisations now conducting post-mortems on their failed technology investments, a consistent finding is emerging: there was no governance. Not weak governance. Not inadequate governance. In many cases, no governance at all. The programmes that consumed tens of millions of pounds operated without stage gates, without independent assurance, without defined decision points, and in some cases without a documented scope or a named sponsor who could be held accountable for outcomes.
We do not have a governance problem because governance failed. We have a governance problem because governance was never there. The dot-com era did not break our oversight frameworks — it revealed that for technology programmes, we had none worth the name.
This is a harder truth than most organisations want to confront. It is easier to blame the market, or the vendors, or the management consultants who encouraged the investment. But the market did not decide to launch these programmes without oversight. The vendors did not decide to skip the business case. The consultants did not decide to bypass the steering committee. Those were decisions made inside the organisation, by people who should have known better and who, in many cases, did know better but chose speed over scrutiny.
Why the Vacuum Existed
The absence of programme governance for technology investments was not an accident of the dot-com era. It was a pre-existing condition that the boom simply made visible.
The Project Management Inheritance
Most organisations have some form of project management discipline. PRINCE2, now in widespread use across the public sector and increasingly in the private sector, provides a structured methodology for managing individual projects. The Project Management Institute’s PMBOK Guide, updated last year, offers a similar body of knowledge from the American tradition. These frameworks are well-established and, when properly applied, effective at the project level.
But a project is not a programme. A project delivers a defined output — a system, a building, a product. A programme delivers a strategic outcome through the coordinated management of multiple projects and the business changes that surround them. The governance that works for a project — a project board, a project manager, a defined scope and plan — does not scale to a programme, where the scope is deliberately fluid, the interdependencies are complex, and the relationship between what is delivered and what the organisation actually gains is indirect and contested.
The discipline of programme management barely exists as a formal practice. The Office of Government Commerce has begun developing a framework — Managing Successful Programmes — but it is in its early stages and adoption is limited. Most organisations that run programmes are doing so with project management tools stretched beyond their design limits, or with no methodology at all.
The Speed Imperative
During the boom, the dominant management philosophy for technology investment was speed. First-mover advantage was the governing principle. The organisations that launched their e-commerce platforms first would capture the market. Those that waited for proper planning, governance, and assurance would lose.
This was not an unreasonable belief at the time — the competitive dynamics of internet markets did appear to reward early entry. But it created an environment in which governance was seen not as a safeguard but as an impediment. Steering committees were too slow. Business cases took too long to prepare. Stage-gate reviews introduced delays that the market would not tolerate.
The result was that many technology programmes of the past three years were launched with what amounted to a verbal agreement between a chief executive and a chief information officer: build it, build it fast, we will sort out the details later. The details — scope definition, benefits identification, risk assessment, stakeholder alignment, resource planning — were deferred indefinitely. In many cases, they were never addressed at all.
The Accountability Diffusion
The third factor is the most structural. Technology programmes during the boom were often sponsored by multiple parts of the organisation simultaneously, with no single point of accountability. An e-commerce platform might be championed by the marketing director, funded by the IT budget, built by an external systems integrator, and operated by a newly created digital business unit. Each party had authority over their piece. Nobody had authority over the whole.
This diffusion of accountability made governance almost impossible, because governance requires someone with the authority to make decisions that affect the entire programme — to change scope, to reallocate resources, to stop work that is not delivering value. When that authority is fragmented across four or five organisational units, each with their own priorities and their own definition of success, the programme drifts. And drift, in the absence of governance, becomes the default mode of operation.
What We Are Learning
The post-mortems now under way are producing a set of findings that, taken together, amount to the beginnings of a governance framework for technology programmes. These findings are not theoretical — they are being extracted, painfully, from the wreckage of programmes that failed.
“Every governance principle we are now articulating could have been written three years ago. The knowledge was available. What was missing was the organisational willingness to apply it when the money was flowing and the pressure was to move fast.”
Finding One — Programmes Need a Single Accountable Owner
The most consistent finding across failed programmes is the absence of a single person with the authority and accountability to make decisions about the programme as a whole. Not a steering committee — committees distribute accountability rather than concentrating it. Not a project manager — project managers lack the organisational authority to make strategic trade-offs. What is needed is a senior business leader — not a technology leader — who owns the outcome that the programme exists to deliver and who has the authority to direct all the resources committed to it.
This role does not yet have a standard name. Some organisations are calling it the programme director. Others use the term senior responsible owner, borrowing from the OGC’s emerging terminology. The label matters less than the principle: one person, accountable to the board, with the authority to say yes, say no, and say stop.
Finding Two — Governance Must Be Designed Before the Programme Starts
The programmes that survived the boom with their credibility intact — and there are some — share a common characteristic: their governance arrangements were defined and agreed before any work began. The decision points, the escalation paths, the criteria for continuation and cancellation, the reporting obligations — all were established at the outset.
This sounds obvious. In practice, it was almost never done during the boom, because defining governance takes time and the imperative was to start immediately. The lesson is that the time invested in governance design is not overhead — it is the mechanism by which the organisation retains control of its investment. Without it, the programme runs on momentum alone, and momentum is not management.
Finding Three — Stage Gates Must Have Teeth
Several organisations had nominal stage-gate processes during the boom but did not enforce them. The gates existed on paper. In practice, programmes passed through them automatically, because nobody wanted to be the person who slowed down the internet strategy. A stage gate that always says yes is not a gate — it is a recording device.
Effective stage gates require three things that were absent in most boom-era programmes:
- Pre-defined criteria that are agreed before the programme starts — not improvised at the gate itself
- Independent assessment by someone who is not part of the programme team and has no incentive to see the programme continue
- A credible stopping option — the organisational willingness to actually cancel a programme that does not meet the criteria, regardless of how much has already been spent
The third of these is the hardest. Sunk cost logic — the reluctance to abandon an investment because of what has already been spent — is the single most powerful force working against effective governance. Overcoming it requires not just a process but a cultural shift: the recognition that stopping a failing programme early is not waste, it is the prevention of greater waste.
Finding Four — Benefits Must Be Defined in Business Terms
The boom-era programmes were justified in technology terms: we need a web platform, we need CRM, we need an e-commerce capability. The assumption was that business benefits would follow from the technology deployment. In almost every failed programme, they did not.
The governance discipline that is emerging requires benefits to be defined before the programme is approved — not in technology terms (“a web-enabled customer interface”) but in business terms (“a fifteen per cent reduction in customer acquisition cost within twelve months of deployment”). The programme’s continuation at each stage gate is then assessed against evidence of progress toward those business outcomes, not against the delivery of technology components.
Building Forward
The governance frameworks that will emerge from this period will not be perfect. They will be built quickly, under pressure, by people who are simultaneously managing the consequences of the programmes that lacked governance. They will be influenced by whatever methodology is closest to hand — PRINCE2, the PMBOK Guide, the OGC’s emerging programme management framework — and they will be adapted, sometimes clumsily, to the specific circumstances of each organisation.
But they will exist. That is the significant change. For the first time, large organisations are recognising that technology programmes require a distinct governance discipline — not project management scaled up, not financial oversight applied sideways, but a purpose-built set of structures and practices for managing complex, multi-project, strategically significant investments.
| Boom-Era Practice | Emerging Governance Discipline |
|---|---|
| No single owner — shared sponsorship | Single accountable owner with board-level authority |
| Governance defined after launch (if at all) | Governance designed and agreed before work begins |
| Stage gates exist on paper, always approve | Stage gates with pre-defined criteria and independent review |
| Benefits assumed from technology delivery | Benefits defined in business terms, tracked at each gate |
| Stopping a programme = failure | Stopping early = sound investment management |
The organisations that build this discipline well will have a lasting advantage. Not because their technology choices will be better — nobody can predict which technologies will succeed — but because they will make and unmake those choices with greater discipline, greater speed, and greater honesty. They will invest with their eyes open, govern with authority, and stop what is not working before the cost of failure becomes the cost of recovery.
The vacuum is real. But vacuums do not persist. Something will fill the space that the dot-com collapse exposed. The question is whether it will be genuine governance — built on accountability, evidence, and the courage to make hard decisions — or merely the appearance of governance, layered on top of the same structural weaknesses that created the vacuum in the first place.