12 Business Case Mistakes Junior PMs Make on Mid-Size Creative Projects
Twelve specific business case mistakes new project managers make on mid-size creative projects — including the single-point-of-failure traps that surface only after a key person leaves.
The twelve omissions that turn a creative business case into a single point of failure
When the business case lives in one person's head, it doesn't survive their vacation, much less their departure.
Junior project managers writing business cases for mid-size creative projects make a recurring set of mistakes that look harmless until someone leaves. The single point of failure pattern is what these mistakes have in common: they each concentrate critical project knowledge in a place — a person, a tool, a private channel — that doesn't survive a transition.
The twelve below are listed in approximate order of severity. Each comes with a one-sentence detective signal: how to spot it before the project loses the person.
- Mistake 1The case lives only in the PM's headDetective signal: when asked to walk through it, the PM speaks fluently but no document captures what they say. Fix: the business case must be a written artifact before any work starts.
- Mistake 2Outcomes defined in agency languageDetective signal: the case uses 'engagement,' 'amplification,' or 'resonance' as success metrics. Fix: replace each with an observable business outcome the CFO would accept.
- Mistake 3Cost based on one quoteDetective signal: only one agency or freelancer was asked to estimate. Fix: at least one comparison quote, even a quick one, and a documented reason for the chosen vendor.
- Mistake 4Critical assets locked in personal accountsDetective signal: the brief, the cost model, or the asset library lives in a personal Google Drive or Dropbox. Fix: every artifact in shared workspaces, named consistently, day one.
- Mistake 5Vendor relationship through one personDetective signal: only one team member has a relationship with the agency. Fix: introduce a backup contact within the first two weeks.
- Mistake 6Approval path lives in chatDetective signal: 'I think Priya approved this in Slack last week.' Fix: every approval has a written confirmation, captured in a shared document.
- Mistake 7Success measured only at end-of-campaignDetective signal: the business case has no mid-campaign checkpoints. Fix: at least two checkpoints, each with criteria for continuing or descoping.
- Mistake 8Risk section names risks but not ownersDetective signal: the risk register has columns for 'risk' and 'mitigation' but not 'owner.' Fix: every risk has a named owner who acknowledges the risk in writing.
- Mistake 9No documented assumptionsDetective signal: when the project shifts, no one can recall what was originally assumed. Fix: an assumptions section in the business case, dated, and reviewed at each checkpoint.
- Mistake 10Founder's verbal commitments not capturedDetective signal: 'the founder said they'd cover X, but it's not anywhere.' Fix: any verbal commitment from leadership is written into the case within 24 hours, with a follow-up email confirming.
- Mistake 11Re-use rights and licensing handled informallyDetective signal: nobody can answer 'can we use this asset for next year's campaign?' Fix: every contract has rights captured in the business case, summarized in plain language.
- Mistake 12No succession plan for the PMDetective signal: if the PM left tomorrow, no one knows where the documents are or who's been promised what. Fix: a one-page handover doc, written on day one and updated weekly.
Quick audit of your current business case
0 / 5- Is every critical artifact in a shared location, not a personal one?
- Does every risk and approval have a named owner with confirmation in writing?
- Could a colleague continue this project tomorrow with what's documented?
- Are mid-campaign checkpoints scheduled with explicit continue/descope criteria?
- Are the vendor relationships dual-contact?
| Mistake | Why it bites | The one-line correction |
|---|---|---|
| Generic cost benefit ratio | Decision-makers ignore unverifiable numbers | Cite the specific cost line and the specific benefit measure |
| Single-stakeholder write-up | Misses the dissenter who later vetoes | Get a second voice into the write before the review |
| No "do nothing" option | Decision looks artificial | Always include the null case with its own cost line |
| Soft success criteria | "Improved engagement" can mean anything | Replace adjectives with thresholds and dates |
| No decision date | The case ages out before sign-off | Print the expiry date on the cover page |
Each of these twelve mistakes is small in isolation. Their combined effect is that the project becomes inseparable from the person running it — which works fine until that person isn't running it. Junior PMs who internalize the twelve checks become the kind of project leads whose work survives them, which is the actual mark of project management seniority. For the related pattern around stakeholder priority collisions, see the priority-collision checklist; for resource-constrained settings, see the construction resource-scarcity quiz.