Three questions that decide if a project fails
Most project failures are set before the contract is signed. Three qualification questions — who signs off, what happens if the date moves, and what the client will do with the work after launch.
Contents
Three questions asked before a proposal predict most project outcomes. Who signs off, and what happens when signers disagree? What does the client do if the date moves? And what will they do with the work after launch? Each maps to a documented failure mode. Ask them in discovery, not in week six.
Key takeaways
- Large IT projects run 45% over budget and deliver 56% less value than predicted, and unclear objectives plus weak stakeholder alignment account for roughly half of overruns (McKinsey and University of Oxford, 2012).
- Inadequate sponsor support is a primary cause of failure for 29% of organizations, and 41% of the lowest performers name it (PMI, February 2018).
- Scope creep affected 52% of projects in PMI's 2018 sample of 5,402 respondents, which makes change control a pre-sale conversation.
- Project size is the strongest predictor available: small projects resolved successfully 61% of the time against 6% for grand projects (Standish Group, 2015).
- Post-launch adoption is a design constraint. Pendo found 80% of features in the average software product are rarely or never used.
Who signs off, and what happens when signers disagree
The first question is not "who is the decision-maker" — everyone answers that with a name. The useful version is: when the marketing director and the commercial director disagree on the homepage, who decides, and how long does that take?
If nobody can answer, the project has no sponsor. PMI's 2018 survey found 29% of organizations naming inadequate sponsor support as a primary cause of failed projects, rising to 41% among the underperforming group. Organizations where more than 80% of projects had actively engaged sponsors reported 40% more successful projects than those below 50%.
In PMI's Pulse of the Profession 2018, based on 5,402 respondents, 29% of organizations named inadequate sponsor support as a primary cause of project failure, and organizations with actively engaged sponsors on more than 80% of projects reported 40% more successful projects. Source: Project Management Institute, 15 February 2018.
Write the answer into the proposal. A named approver with a stated response window is a contractual fact, not a courtesy.
What happens if the date moves
Ask what breaks if launch slips by three weeks. The answers separate two very different projects. If the reply is a trade show, a regulatory deadline or a campaign already bought, the date is fixed and scope must flex. If the reply is vague, the date is a preference and scope can be held.
Getting this backwards is expensive. McKinsey and the University of Oxford examined more than 5,400 IT projects with initial budgets above $15 million USD. On average those projects ran 45% over budget and 7% over schedule while delivering 56% less value than predicted, and each additional year added a further 15% of cost overrun.
Size is the lever you control. The CHAOS Report 2015 resolved small projects successfully 61% of the time and grand projects 6% of the time. Splitting one large engagement into three sequenced ones changes the odds before any technology decision is made. Our position that scope, not hours, is the right unit to quote exists because it forces this conversation early.
What the client will do with the work after launch
The third question is the one most often skipped: who runs this in six months, with what budget, and what will they change first? A website nobody is resourced to update is a depreciating asset from the day it ships.
Pendo analyzed feature usage across 615 subscriptions and reported that 80% of features in the average software product are rarely or never used, with publicly traded cloud companies investing up to $29.5 billion USD building them. Source: Pendo, 5 February 2019.
The same pattern shows up in brand and web work. Sections get commissioned because someone in the room wanted them, then never get content. Asking who owns each area after launch removes scope more reliably than any prioritization workshop, and it produces the named ownership that keeps documented process alive.
How to ask before you write the proposal
These questions belong in discovery, alongside the ones that replace a deliverable-shaped brief with a problem-shaped one. Ask for the disagreement scenario, not the org chart. Ask what the date protects, not whether the date is firm. Ask what happens in month six, not whether the client is committed.
Then price the answers. A project with an unnamed approver, a soft date and no post-launch owner is not cheaper — it is different. It should carry a change-control clause, a shorter first phase and a milestone structure that protects collection as much as delivery.
FAQ
What if the client refuses to answer these questions?
Refusal is itself an answer. A buyer who cannot name an approver or a post-launch owner is describing an organization that has not decided to do the work yet. Reduce the first phase to a paid discovery engagement, deliver a decision document, and re-quote the build afterwards.
Are these questions relevant for small projects too?
Yes, and they take fifteen minutes. The Standish Group's 2015 data shows small projects succeed far more often than large ones, but the failure causes are identical — unclear approval, drifting dates and no owner. The cost of asking does not scale with project size.
How do I raise change control without sounding defensive?
Frame it as a pricing mechanism rather than a legal one. Explain that 52% of projects in PMI's 2018 sample experienced scope creep, that you assume changes will happen, and that the clause exists so a change costs a re-quote rather than an argument.
Do these questions replace a discovery phase?
No. They qualify whether discovery is worth running and at what depth. Discovery answers what to build; these three questions establish whether anyone can approve it, whether the timeline is real, and whether the result will be maintained.
What to do with the answers
Ask all three before you quote, write the answers into the proposal in the client's own words, and let them set the phase structure rather than the price. Where an answer is missing, shorten the first phase instead of discounting it. Six months out, the projects that held their scope will be the ones where a named person answered the disagreement question on day one, and that correlation is worth tracking across your own pipeline.
