Decision Analysis In Project Management

90% of teams don’t fail because of lack of talent. They fail because nobody knows who owns what. I’ve seen strong teams with senior people and clear goals still miss deadlines, clash on decisions, and redo the same work twice. Not a skills problem. A clarity problem. Two things were always mixed together: execution and decision-making. So I separated them. RACI → for execution clarity Who does the work Who owns the outcome Who gives input Who stays informed DACI → for decision clarity Who drives the process Who makes the final call Who contributes Who gets informed What changed: • Meetings got shorter • Decisions got faster • Accountability became obvious • “I thought you owned this” disappeared The key insight: RACI fixes execution chaos. DACI fixes decision chaos. Most teams try to solve both with one framework and end up with neither. What I do now: – RACI at the start of every project – DACI for every meaningful decision – No shared ownership – No shared final approval Clarity scales. Ambiguity kills velocity. If your team feels busy but stuck this is probably what’s missing. – Natan Mohart

Uncertainty isn’t the enemy of leadership. Silence in uncertainty is. Markets shift. Geopolitics flare. Technology disrupts. No leader can predict exactly what comes next. The mistake isn’t saying “I don’t know.” The mistake is leaving it there. Silence creates space for fear. Scenarios create space for confidence. The leaders I know say this: “We don’t know the future…But here are three ways it could play out, and here’s how we’ll respond to each.” That shift replaces anxiety with structure. Here’s how scenarios guide decisions: 1. Best Case → Maximise Opportunity • If growth rebounds, be ready to scale • Line up resources and move first • Optimism matters only if you’re prepared 2. Base Case → Navigate Steady State • In uneven recovery discipline wins • Tier your investments • Forecast cash tightly • Normalise quarterly adjustments 3. Worst Case → Build Resilience • Protect non-negotiables • Pre-approve cost levers • Over-communicate with empathy, reinforce purpose • Trust is forged in downturns, not booms. The real power is in cascading this skill to teams: → Model vulnerability (“I don’t know yet”) → Teach them to sketch 3 scenarios in 15 minutes → Anchor every path to concrete actions → Repeat until it becomes part of culture At 6 months, fear gives way to clarity. At 2 years, resilience becomes second nature. Remember, great leaders don’t eliminate uncertainty. They equip their people to move confidently within it. That’s how you scale trust, resilience, and momentum, inside your company and across your partnerships. --------------------------- Avoid missing insights like this. Get cheatsheets like this each Wednesday. Subscribe to my free newsletter: https://philhsc.com ➕ Follow me, Phil Hayes-St Clair for more like this.

Decision-dependent uncertainty The prestigious journal Mathematical Programming Series B just announced they were soliciting papers for  stochastic programming with “decision-dependent uncertainties.”  For the uninitiated, the field of stochastic programming likes to generate samples of “scenarios” of what might happen in the future - these are represented as “scenario trees” that are then used to plan for the future to make a decision now.  This is an idea that dates back to a 1955 paper by George Dantzig who invented the simplex method.  There are *thousands* of papers on this topic written every year, but I like to repeat the question asked of me by Ed Rothberg (he is the “Ro” in the Gurobi optimization library): “Warren, does anyone actually use stochastic programming?” Scenario trees have to be generated in advance, which means that the random events do not reflect what decisions have been made.  Even with this simplification, stochastic programming (with scenario trees) creates problems that are much harder than the more familiar deterministic lookahead policies, which is a reason why this approach is rarely used. Now they want to address the problem of making the uncertainty depend on previous decisions.  One example of where this happens is in truckload trucking, where the appearance of random loads to be moved might depend on whether a truck is in a region. There are several ways to handle decision-dependent uncertainty: Use a parameterized, deterministic lookahead - I call this a “cost function approximation” … start with a deterministic lookahead, then introduce parameters to make the deterministic solution more robust (schedule slack, buffer stocks, discounts on uncertain forecasts).  Finally, tune the parameters in a stochastic simulator that captures (if you wish) state or decision-dependent uncertainties.  The tuning of the parameters picks up this behavior.  See https://lnkd.in/eEcpM4Ex for an introduction to this approach, and chapter 13 (or section 19.6) of https://lnkd.in/dB99tHtM for a more thorough description. If you really need a stochastic lookahead and want to have state- or decision-dependent uncertainties, consider stochastic lookaheads that use simulation, such as ADP-based methods (section 19.7 of https://lnkd.in/dB99tHtM for a discussion of stochastic lookaheads). Monte Carlo tree search (section 19.8) solves a full stochastic lookahead where it is quite easy to introduce state or decision-dependent uncertainties, but MCTS only works for scalar decisions. Take a look at optimistic MCTS in section 19.8.4. Make sure you evaluate your policy in a proper simulator.  More sophisticated lookahead models do not always translate to better policies.    I strongly recommend making a parameterized deterministic lookahead your base policy for comparison (strategy 1 above).  Your results will be highly problem-dependent, so make sure you have an application where this really matters.

Most “sensitivity analyses” aren’t sensitive at all. They’re just a few +/- tweaks in Excel. If you want to do it properly, here are 5 quick tips: 1. Pick the right drivers → Focus on the 3–5 variables that truly move the business (price, volume, churn, CAC). 2. Test extremes, not just margins → Push assumptions until the model breaks; that’s where you find the risks. 3. Use scenarios, not scatter → Structure downside, base, and upside cases with clear triggers. 4. Visualize impact → Tornado charts, spider plots, or even simple waterfall views make risks tangible for leaders. 5. Connect to decisions → End every sensitivity test with: “If X happens, here’s what we’ll do.” Sensitivity analysis isn’t about proving your model. It’s about showing leaders where it bends, and where it breaks. P.S. What’s the most surprising variable you’ve seen sink a “bulletproof” plan?

Is your team tapping into collective wisdom or falling into groupthink? 🤔 🫶🏼 Groupthink occurs when a group's desire for harmony and agreement causes members to ignore different opinions, avoid critical thinking, and make poor decisions just to keep the peace. ☝🏼Collective wisdom happens when the aggregated opinions, knowledge or predictions of a diverse and independent group of people leads to more accurate decisions. To shift a team from groupthink to collective wisdom, the decision-making process should be structured to encourage open communication, critical thinking, and the value of diverse perspectives. How to facilitate this shift? 📝 Individual pre-work: Ask members to independently analyze the issue and prepare their opinions before group discussions. This can help prevent initial ideas from dominating the conversation. 😈 Use rotating roles ... such as "devil's advocate," "fact-checker," and "process observer" to various members, rotating these roles to ensure balanced participation and a critical examination of the group's decisions. 🧠 Use brainwriting instead of brainstorming So the ideas can get first generated individually, then shared and discussed as a group What methods have you found effective in encouraging independent thinking and open dialogue in group settings?

An SVP told me she had to present a major strategy shift to her board, and she knew they weren't going to like it. She ran the numbers. She explored alternatives. She knew this was the right decision. But she also knew the board would push back…hard. They'd question her assumptions. They'd challenge her math. They'd ask "why not do this instead?" And she needed to be ready for every question—not defensive, but confident. Here's what she did: Instead of simply rehearsing her talking points over and over, she used AI to simulate the board conversation before it happened. She gave AI the context:  👉 Her proposal (the strategy shift she was recommending) 👉 The board composition (CFO focused on cash flow, tech founder focused on innovation, nonprofit leader focused on mission) 👉 Her reasoning (why the change in direction, why now, what she'd already considered) Then she asked: "Given what you know about this board and this proposal, what are the 10 toughest questions they'll ask me? What objections will they raise? Where will they think my logic is weak?" AI gave her 12 questions she hadn't fully thought through. Questions like: → "How do we know this won't hurt our competitive position?" → "What's your contingency plan if revenue doesn't recover as projected?" → "Why these departments and not others?" She wasn't ready for most of them. So she spent the next two days pressure-testing her answers. Not writing a script; but thinking through her reasoning so she could speak confidently no matter how the conversation went. When she walked into the board meeting and those questions inevitably came up, she answered them with clarity, data and confidence instead of scrambling. One board member even complimented her on how well prepared her presentation was. Here's what made the difference: ✅ She didn't use AI to write her presentation and run with that. She used AI to think through the conversation she was about to have. ✅ She didn't ask AI for answers. She asked AI to help her find the gaps in her own thinking. ✅ She prepared for the questions, not just the pitch. That's strategic AI use. Have a high-stakes presentation coming up? Don't just rehearse what you want to say. Simulate the conversation. Pressure-test your thinking. Be ready for the questions you haven't considered yet. ⭐ BONUS TIP: If you’re using ChatGPT, you can tap the “Use Voice” feature and actually practice answering the questions aloud with a responsive voice in real-time. Give it a try and let me know how it goes!

When faced with new policies, firms must decide whether to "wait and see" or invest in new #technologies and processes to comply. But how can companies make informed decisions when #policy implementation is uncertain? My latest research with Eun-Hee Kim and Maggie Zhou offers insights! We found that firms can gauge policy implementation commitment by analyzing communication exchanges between regulatory agencies and policymakers. Costlier agency communication signals stronger future policy implementation, encouraging firms to invest in long-term technologies. Empirically, we delved into the early years of the European Union's Emissions Trading Scheme (EU-ETS), a policy aimed at reducing #greenhouse gas emissions. We examined the European electric power investments in #renewable energy facilities (2004-2009) in response to country-level agency communications to the EU Commission. During this period, #carbon allowances were freely given away, #emission caps were not binding, and it was unclear whether country agencies would strictly implement the #EU-ETS. These, combined with the plunging price of carbon during the 2007–2009 financial crisis, cast doubt on any concrete, short-term return on emissions-reducing investments. Read our full paper to learn how firms can strategically respond to policy implementation uncertainty. https://lnkd.in/guzzc7Fz #policyuncertainty #strategy #sustainability #energy #regulation USC Marshall School of Business

I don't know what I don't know - a common challenge that can derail projects and team success. Having led multiple teams and projects across Asia Pacific, I've learned that addressing unknown unknowns is crucial for project success. Here's how I approach this challenge: 🔍 Start with structured discovery sessions. I always kick off projects with comprehensive discovery workshops where team members can openly share their knowledge gaps and concerns. This creates psychological safety and helps surface potential blind spots early. 📊 Map out knowledge domains. I try to identify different areas of expertise needed for the project - technical, business, regulatory, market-specific requirements. This helps highlight where we might have gaps in our collective knowledge. 🤝 Engage subject matter experts early. When dealing with new markets or technologies, I proactively bring in experts from different functions or external consultants. Their insights often reveal critical considerations we hadn't thought about. Along the way, I will proactively consult them for issues that crop up along the way too. ❓ Ask better questions. I've learned that asking the right questions is more important than having immediate answers. Some key questions I always ask: - What regulatory or compliance issues might we face? - What market-specific factors should we consider? - What similar projects have we done before? - What were the unexpected challenges? 🔄 Regular retrospectives. I schedule frequent check-ins where teams can safely discuss new uncertainties that emerge. This creates a culture of continuous learning and adaptation. 💡 Build in buffer time. When planning projects, I always account for the "unknown unknowns" by adding contingency time and budget. The more complex, the more likely chance of delays. This has saved many projects from delays when unexpected challenges arose. So, fellow leaders and project managers, how do you handle the "unknown unknowns" in your projects? What strategies have worked well for you in identifying and addressing knowledge gaps? #leadership #coaching #strategy #jenelim

Mitigating Liquidity Risk: Key Tactics for Banking Treasury Management In the dynamic landscape of banking, liquidity risk stands as one of the paramount challenges for treasury management. It is not merely about having sufficient cash on hand but rather about ensuring the availability of liquidity when and where it's needed most. As such, mastering liquidity risk management is essential for the prudent functioning of any banking institution. Understanding the nuances of liquidity risk is the first step towards effective mitigation. It encompasses the risk of being unable to meet financial obligations as they come due without incurring unacceptable losses. This can arise from funding mismatches, unexpected deposit withdrawals, or disruptions in the interbank lending market. However, merely grasping the concept is not enough; proactive measures must be taken to mitigate this risk. Here are some key tactics that banking treasuries can employ: 1. Stress Testing Scenarios: Conducting rigorous stress tests to simulate adverse market conditions can provide valuable insights into potential liquidity shortfalls. By analyzing various scenarios, treasuries can identify vulnerabilities and develop contingency plans accordingly. 2. Diversification of Funding Sources: Relying too heavily on any single funding source can expose a bank to significant liquidity risk. Diversifying funding sources, including wholesale funding, retail deposits, and access to central bank facilities, can enhance resilience against funding disruptions. 3. Maintaining High-Quality Liquid Assets (HQLA): Holding a portfolio of high-quality liquid assets, such as government securities and cash reserves, serves as a buffer during periods of liquidity stress. Ensuring sufficient HQLA levels relative to funding needs is a prudent risk management practice. 4. Establishing Contingency Funding Plans (CFP): Developing robust contingency funding plans that outline strategies for accessing liquidity in emergencies is essential. These plans should outline clear escalation procedures and specify the roles and responsibilities of key stakeholders. 5. Monitoring and Early Warning Systems: Implementing robust monitoring mechanisms and early warning systems enables treasuries to detect liquidity risks in real-time. By closely monitoring liquidity metrics and market developments, banks can take timely corrective actions to mitigate potential threats. In conclusion, effective liquidity risk management is indispensable for the long-term viability of banking institutions. By understanding the nature of liquidity risk and implementing proactive risk mitigation strategies, treasuries can safeguard against potential liquidity shocks and ensure the uninterrupted provision of financial services. #Banking #TreasuryManagement #RiskMitigation #LiquidityRisk #FinanceManagement

The #1 mistake I see companies make isn’t strategy. It’s collaboration. And it’s costing you more than you think. Here’s what’s happening: You’re treating performance like an individual sport. One “expert.” One “owner.” One “hero.” But here’s the problem with that approach… One study of 222 project teams found something remarkable: Groups outperformed their most proficient member 97% of the time. 97%. Let that sink in. So if your team is stuck, stop asking: “Who’s the expert?” Start asking: “How do we solve this together?” That shift changes everything. Here’s the simplest question I use with leadership teams: “What do you see that I don’t?” Then I do one thing most teams skip… I make it safe for the quiet people to speak first. Why? Because the first voices shape the whole room. If the loudest person goes first, everyone else just agrees or stays silent. Equal voice isn’t a “nice-to-have.” It’s how you get: • Better decisions • Faster alignment • Fewer blind spots And you have to measure it—not just outcomes, but collaboration behaviors. Try this for 30 days: Every meeting ends with: “What’s one risk we’re not naming?” “What’s one idea we didn’t hear yet?” Then assign the next step as a team— not a hero. Here’s the truth: If you fix collaboration, results follow. If you don’t, even great strategy collapses under friction. Stop building teams that depend on one genius. Build teams that compound intelligence. That’s how you turn good teams into unstoppable ones.