"The objectives are formulated in broad, aspirational terms and are not accompanied by specific, measurable success criteria; the panel is therefore unable to assess the extent to which they would be achieved." What they were really penalising Not your ambition: its verifiability. Under Excellence, the evaluator scores your objectives against one quiet test: if this project ends, how will we know it delivered? Phrases like "advance understanding," "develop a novel framework," or "significantly improve" read as intentions, not commitments. With no baseline, target, or metric, the panel has nothing to check delivery against. So the objective caps your score, however exciting the science underneath it. The fix - Make each objective SMART: state the what, the how much, and the by when. Pair every verb with a number and a deadline. - Give the evaluator a success criterion they can tick: baseline → target ("from X to Y," "validated in N models," "≥ threshold"). - Separate the aspiration (your vision) from the objective (the measurable step that serves it). Keep both: just label them clearly. - Sanity-check: could a non-specialist confirm each objective was met from your own KPIs alone? If not, tighten it. AI will draft objectives fast; the evaluator still scores them against the criterion. Speed doesn't make an objective measurable. Structure does. Deep-dive in Course 3 — Excellence, Ambition & Impact. Free companion patterns at scientificmanagers.eu. (Evaluator lines in this series are composite and illustrative of common evaluation patterns — not quotations from any specific proposal.)

"The expected impacts are described as desirable outcomes rather than a credible pathway to achieve them; the mechanisms, target groups and the assumptions on which uptake depends are not sufficiently substantiated, and the proposed measures to maximise impact remain generic." What they were really penalising Not your ambition. Your machinery. Impact is scored on plausibility, not on how much good you promise. When you write "this will benefit patients / industry / policy," the evaluator reads a destination with no road. They are trained to ask: through whom, by what step, and what has to be true for it to work? A pathway that skips the intermediate actors, the measurable change, and the dependencies reads as a wish list. Wish lists lose points even when the science is strong. The fix - Write the chain explicitly: result → who adopts it → what they do differently → what changes, measured how. One line per link. - Name the target group and the specific measure to reach them; "dissemination to stakeholders" is not a measure. - Surface the assumptions and dependencies (regulatory, behavioural, commercial) and say what you do if one fails. - Attach an indicator and a baseline to each claimed impact so it can be verified, not just believed. Impact is where strong science quietly loses points. Course 3 (Excellence, Ambition & Impact) rebuilds the pathway the way evaluators score it. Start at scientificmanagers.eu. (Evaluator lines in this series are composite and illustrative of common evaluation patterns — not quotations from any specific proposal.)

The autumn deadlines are closing. If you've hit submit, first, well done, that's the hard part most people never finish. This one's about what comes next. While you wait (evaluation and information on results typically take some months, check the timeline in your call), here's what's actually worth doing: 1. Save everything, cleanly. Your final submitted PDF, the annexes, the exact call version, the internal review notes. When feedback arrives you'll want to compare it against what you actually sent, memory won't be reliable by then. 2. Write down, now, the parts you already know were weak. Do it while it's fresh. This becomes gold if you need to resubmit, and it separates your honest self-assessment from the evaluators' view so you can tell where they agree with you and where they saw something you didn't. 3. Don't wait idle. If there's a sensible next call for this idea, sketch how you'd adapt it. Many funded proposals are second or third attempts, treating the first submission as the end is the mistake. 4. Rest a little. Deadline season is a sprint and the next one comes around. When the Evaluation Summary Report arrives, read it like an engineer, not a defendant: - Read it twice. First pass you'll feel the criticism, second pass you'll actually absorb it. Don't reply to it emotionally, from anyone. - Map each comment to the criterion it sits under. A comment under Impact is telling you something structural about your impact section, not a matter of taste. - Separate the fixable from the framing. Some comments point at a genuine gap you can close next time, others reflect how competitive the call was. Sort them, and spend your energy on the fixable. - Look for what they did NOT flag, too. The parts they passed over quietly are usually your solid ground to build on. - If you plan to resubmit, turn the report into a concrete revision list, comment by comment, addressed or consciously not. That list is the most valuable thing the process gives you, more than the score.

Win of the week first, because in deadline season the small ones keep you going. If you shipped something this week, a submitted proposal, a finished work plan, a partner who finally sent their part, a section you'd been avoiding, post it below. It counts. A submitted-but-imperfect proposal beats a perfect unsubmitted one every single time, so "I hit send" is a real win and I want to see it. Now one small rigour nugget you can apply the moment you're back in your draft, the vague-to-measurable objective. It's the single most common weakness I see, and one of the fastest to fix. Before: "Objective 3: Improve the performance of the system." Why it fails: improve from what, to what, measured how, by when? A reviewer can't score it because there's no way to tell, at the end of the project, whether you did it. An unmeasurable objective drags down both your excellence and your credibility, because it signals you haven't defined success. After: "Objective 3: Reduce the system's energy use per processed sample by at least 30% versus our current baseline (Method X, 2025), verified on the three reference datasets in WP4 by month 30." What changed: a direction and a target (30% reduction), a baseline (so the target means something), a verification method (the reference datasets), and a deadline (month 30). Now it's checkable, and a checkable objective reads as a team that knows exactly what it's doing. You don't need heroic numbers. You need a target, a baseline, a way to verify, and a time. Even one of those, added to a vague objective, moves it forward. Reply with your win. And if you're brave, paste one objective and I'll help you make it measurable, right there in the thread.

I added something to Funding Radar (fundingradar.eu) this month that I had wanted for a long time, and the result surprised me enough to be worth sharing. You can now enter several organisations at once instead of one. It builds a research profile for each from public publication records, combines them, and ranks open calls against the group. Testing it, I put in three institutions of very different sizes: 6,901 indexed publications, 1,321, and 73. Between them they covered 62 research themes. The smallest of the three held twelve of those themes on its own. Neither of the bigger two published on them at all. That is the thing I keep coming back to. In twenty years of proposal work I have watched small partners get talked out of consortia because they looked minor next to a large university, and here was the arithmetic showing why that reasoning is wrong. Volume and contribution are not the same measurement. The part I find most useful is what it says about a specific call: which of that call's themes the group can evidence, and which nobody covers. A call can score well and still be mostly outside what the group does: the subject is right, the people are missing. That is a partner problem, not a fit problem, and I have seen the two confused often enough to want the distinction on screen. It has one honest limitation I want to state rather than bury. The profiles come from published research, so companies, end-user organisations and patient groups do not appear, however essential they are to a Horizon consortium. Where the Commission's records show what was actually funded under a topic, the tool reports which kinds of organisation those consortia contained. It cannot find you a company. It can tell you that you are missing one. Free, no account: https://fundingradar.eu/consortium