Why Most Idea Lists Stay Useless
You've probably compiled a list of ten project concepts, feature requests, or startup ideas and then just... dropped them. They sit in a doc somewhere until the next brainstorm cycle. This is where the Ideas Top 10 framework actually matters. It's not a scoring matrix or a fancy weighted decision tool. It's a filtering discipline that forces you to eliminate ideas before you ever try to execute on them. Here's the actual process. Generate ideas freely first, without judgment. Get thirty to fifty out. Then apply five specific criteria to each one, and knock out anything that fails more than two. What survives becomes your Ideas Top 10 list. The five criteria are straightforward enough that you can run this in a single sitting: Impact potential — How much does this move the needle if it works? Rate it 1 to 5. Don't guess at revenue numbers. Think about whether this changes a fundamental constraint or just adds a minor feature on top of existing work.
Execution feasibility — Can you actually build this with the resources you have right now? This is where most people lie to themselves. I've seen entire teams rate an idea a 4 on impact and a 5 on feasibility when the idea required a proprietary dataset they didn't have access to. That kills the idea immediately. Time to value — How long until someone gets something useful out of this? A six-month build with no interim milestones is a 2. A prototype users can test in two weeks is a 4 or 5. This criterion alone has saved me from pursuing three ideas that looked brilliant on paper but would have consumed eighteen months with nothing to show. Originality — Is there meaningful differentiation from what already exists? This doesn't mean your idea has to be revolutionary. It just means you're not recreating something that already has twenty competitors with bigger budgets. A 1 here means "I saw this same thing last week." A 5 means "nobody is doing this in a directly comparable way."
Alignment — Does this fit your actual goals, or just sound impressive? This is the criterion most people skip. I used to skip it too, which is how I spent four months building something that nobody in my org actually wanted. Rate your current objectives honestly. If none of them align, the idea gets cut regardless of the other scores. After scoring, any idea with two or more ratings below 3 gets eliminated. The remaining ideas, usually between six and twelve, form your Ideas Top 10. If you have fewer than six, your initial generation phase was too conservative and you need another round. If you have fifteen, your elimination threshold is too lenient — tighten it.
Get the Full Details
What People Get Wrong About This Process
The biggest mistake I see is treating the scoring as objective when it's actually speculative. You're rating ideas against futures that haven't happened. The framework works precisely because it makes the speculation visible instead of hiding it behind a gut feeling. When two people score the same idea differently, you've found a disagreement about assumptions you can now test instead of a disagreement about taste. Another common failure mode is applying the criteria retrospectively. People look at an idea they've already attached themselves to and start inflating the scores to justify keeping it. I caught myself doing this once with a product feature I'd been thinking about for weeks. I went back through and lowered the feasibility score by two points after admitting I'd never actually validated the core technical assumption. The idea should have died. I kept it anyway and spent three months confirming my worst fear. The method also breaks down when applied to ideas that operate on different timeframes. A ten-year research goal and a two-week marketing campaign will never land in the same Ideas Top 10 fairly. Keep your lists thematically tight. Sort ideas by category before you score them if they span wildly different horizons.
A Real Example From My Own Work
Last year I ran this framework on a list of twelve internal tool suggestions from our engineering team. The raw generation phase produced about forty items across five categories. After the first cut, I had eighteen. The second pass brought it down to nine. One idea that scored a 5 on impact and a 1 on feasibility turned out to require a database migration we couldn't complete without taking the production system offline for an estimated three days. We thought we could avoid the downtime with a dual-write strategy, but the feasibility score of 1 should have been a warning sign. Nobody on the team had actually attempted that pattern before. The idea got cut and we moved on. The winning idea from that session scored modestly everywhere — a 3 on impact, a 4 on feasibility, a 5 on time to value, a 3 on originality, and a 5 on alignment. It wasn't exciting. It was a reporting dashboard that automated a manual weekly export process. It took two weeks to build. It saved the operations team about eleven hours per week. That's the pattern these frameworks surface: the boring ideas that compound over time tend to outperform the dramatic ones that stall out. There's no download required for this. It doesn't live in any software. You need a spreadsheet, a stopwatch for the timing estimates, and the willingness to kill ideas you spent time thinking about. The hardest part isn't the scoring. It's the deletion. But that's the whole point.
