A tool for optimizing how you spend your time across multiple projects by allocating a fixed resource (time, money, focus) across a set of proposed projects.
As a medical student, I constantly juggle research projects, clinical learning, personal projects, and skill development. Some projects demand months of effort before yielding results, while others provide quick wins. Some maintain their value even at 90% completion, while others lose all value if you don't push through to the end.
Traditional project management treats all projects the same - you either do them or you don't. But that's not how real projects work. I needed a better way to decide not just what to work on, but how much effort to put into each project.
I've always been fascinated by the Pareto Principle - the idea that 80% of results come from 20% of effort. But as I applied it to my projects, I realized it's not that simple. Different projects have wildly different Pareto "dynamics." Some projects give me 80% of the value with just 10% effort (like resubmitting an abstract to multiple conferences). Others might need 60% effort to reach that 80% value (like systematic reviews).
At first, I thought I could capture this with a single number - what percentage of effort gets you to the "good enough" point? I called it the Pareto Index, representing where the efficiency inflection point occurs. A project with a PI of 20% peaks early (quick win), while a PI of 80% means a long slog before payoff.
But then I hit a problem. Two projects might both follow a 20/80 pattern, yet feel completely different to work on. Why? Because the shape of the curve matters. One project might give me 60% value at just 10% effort (fantastic!), while another gives only 20% value at that same 10% effort mark. Same inflection point, totally different efficiency journey.
That's when I realized I needed to model the entire curve. Not just where it peaks, but how steeply it rises, how sharply it falls, whether it plateaus. A project that ramps up quickly lets you capture value early. A project with a gentle decline forgives you for pushing past the peak. A project that plateaus before its peak is telling you to stop early.
This led to the full framework you see here - multiple parameters that together capture a project's complete efficiency profile. It's still rooted in the Pareto Principle, but acknowledges that real projects are more nuanced than any single ratio can express.
Every project has an efficiency curve - how much value you get per hour changes as you progress through the project. A case report might give you great returns for the first 10 hours, then sharply diminish. Original research might require 40 hours of groundwork before becoming productive, but then sustain high efficiency for months.
The key realization: you don't have to complete every project to 100%. Often, stopping at 70% or 80% captures most of the value while freeing up time for other high-impact work. The challenge is finding out how much to pour into the project. This isn't a very accurate tool, but to help estimate project value along the lifecycle and compare several projects it should be useful. It's like a weight scale, even if your scale you step on is a little off, the relative weights of different objects you put on it should be consistent, and you can evaluate your weight change over time (i.e., your confidence in the change is higher than your confidence in the absolute value).
The framework models each project using six parameters (for now) that capture its efficiency journey:
Activation Energy - Some projects have a "dead zone" where you invest time but see no returns yet. Think of learning a new programming language or starting a research collaboration.
Peak Timing (Pareto Index) - When does the project hit maximum efficiency? Quick wins peak early, marathons peak late.
Ramp and Decline Rates - How fast does efficiency grow? How sharply does it fall? These rates determine whether a project is forgiving or requires precise timing.
Peak Value - At its best, how efficient is this project? Some projects at their peak might deliver 10x returns, others only 2x.
Total Time - How many hours would it take to fully complete this project? This scales everything to real time.
Start by modeling a project you know well. Maybe that literature review you just finished, or that side project you've been working on. Play with the parameters until the curves match your experience - when did it feel most productive? When did you start hitting diminishing returns?
Once you've calibrated your intuition, add your current projects to the portfolio. Be honest about the parameters. That research project with the famous professor might have high peak value, but also high activation energy.
The optimization algorithm then allocates your time in small increments, always choosing the project with the highest current efficiency. It knows to push through activation periods, respect your time constraints, and stop projects when efficiency drops below what you could get elsewhere.
While I built this for my own needs, the framework applies anywhere you're juggling multiple projects. Software developers balancing features and technical debt. Academics managing research, teaching, and service. Entrepreneurs deciding how to split time between product, marketing, and fundraising.
The math works the same whether you're optimizing research papers or product features. What matters is that you're choosing between projects with different efficiency characteristics, and you have limited time.
If you're interested in the mathematical details, click the "Theory & Guide" button in the tool. The framework uses parametric curves to model efficiency, with a greedy optimization algorithm modified to handle activation costs and diversity constraints.
I'm exploring a potential paper on this, possibly exploring extensions like value persistence (how long returns last after effort stops) and uncertainty quantification. But the core framework is useful today, even in its current form as a back-of-the-napkin style calculator.
Just open index.html in your browser. No installation needed. Start with the preset project types to get a feel for the parameters, then model your own projects. The visual feedback makes it intuitive - you'll see immediately how changing parameters affects the efficiency curves. you can also try it here.
This is version 1.0, focused on the core optimization problem. I'm planning to add:
- Historical tracking to validate predictions
- Team collaboration features
- Integration with time tracking tools
- Machine learning to predict parameters from project descriptions
If you find this useful, I'd love to hear about your experience. What patterns do you see in your projects? What features would make this more useful? Feel free to open issues, submit pull requests, or just reach out.
MIT License - use this however you want.
Built out of necessity by a busy bee who needed a better way to manage too many interesting projects.