Anyone who makes a promise or a demand should say later whether it came true. We apply this to the sanctions we write about, and to ourselves. This page explains what that means in practice and where you can check whether we keep to it.
In two minutes
- Beforehand, we write down what we expect. Afterwards, we show what happened, even when the result is poor.
- Anyone who writes to us gets a reply.
- We test announcements by governments and authorities with the same three questions we apply to our own.
- We mark evidence, assessments and opinions separately.
- Mistakes go on a public corrections list. After the week we disclose what came in and what it was spent on.
The basic idea
If the garage says your car will be ready on Friday, you expect the car on Friday, or a phone call. If neither comes, you trust them less next time and plan differently. That is not a theory; it is everyday life.
Our working assumption follows from this: people pull back when what they are promised, what they can actually do and what they learn afterwards no longer fit together. If you never find out whether something was worth it, at some point you stop trying. This is an observation from our work, not a proven rule.
Often only the form remains. There is a confirmation, a report, a press release, but nobody checks whether what was promised actually happened. A receipt is not yet a result.
So the question we ask to be measured by is this: do expectations learn from results? Anyone who expected something should be able to see afterwards whether it happened, and adjust their next expectation accordingly. That takes two things: the expectation is fixed in advance, and afterwards someone reports back.
Six commitments
- Write it down first, compare afterwards. Before the week of 16 to 24 November we write down what we expect: how many people will come, how many of our enquiries will be answered, how much the aid organisations will receive, whether the costs will be covered. For each expectation we say how confident we are. After the week we enter what actually happened next to it, and what we will do differently next time. After the cut-off date we no longer change the expectations; if we ever do, the old version stays visible with its date.How to check: on the feedback page, outlined below.
- Every enquiry gets a reply. Anyone who writes to us gets a reply, as a rule within ten working days, even if the answer is "no" or "we don't know yet". During the week itself it may take longer; if so, we say so. Advertising and abusive messages are excluded. It also works the other way round: anyone who has helped us or given us something will hear from us what came of it.How to check: the feedback page shows how many enquiries came in and how many we answered within ten working days.
- The same standard for policy. With sanctions and blockades we do not ask first whether they are good or bad, but: What goal was announced? What happened? Who checked? An example from the United States: the 1996 law known after its sponsors as the Helms-Burton Act lists among its purposes encouraging free and fair democratic elections in Cuba. An example from Europe: the 1996 EU Blocking Statute is meant to protect companies in the EU from the effects of foreign sanctions laws; the German federal government says it has no figures on how many companies in Germany were affected in the past ten years. We put the same three questions to reforms announced by the Cuban government, to decisions of the United Nations and to ourselves.How to check: our texts state the goal, the result and who checked it, with sources. Where no one has checked, we say so.
- Evidence, assessment and opinion kept apart. The box below this list explains what that means. In new texts we separate the three visibly; older texts will be converted step by step.How to check: by the labels in the text and the source list at the end.
- Mistakes go on the corrections list. When we get something wrong, we do not change it quietly. The corrections list says what was wrong, what applies now, since when, and, if the person agrees, who pointed it out.How to check: in the corrections list, linked at the foot of every page.
- Money in the open. Donations for the people of Cuba go directly to the aid organisations, not to us. There is a separate fundraiser for the costs of the week. After the week we disclose what came in and what it was spent on; people who contribute 1,000 euros or more are named only with their consent, smaller contributions only on request.How to check: on the pages Donations and Fundraiser and in the accounts after the week.
- EvidenceA statement with a source and a date that you can check yourself.
- AssessmentOur reading of the evidence. We say how confident we are and what would change our mind.
- OpinionA stance, for example in a speech on stage. It is marked as opinion and does not replace evidence.
The 2026 feedback page
This is what the feedback page for the 2026 week will look like. We enter our expectations before the week begins and the results after 24 November. Until then the fields stay empty. If a result is worse than expected, it appears there just as a good one would.
| What | Our expectation | How confident | Result | What we learn |
|---|---|---|---|---|
| Visitors, all days | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Tickets sold | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Our enquiries to institutions, media and sponsors: how many answered | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Enquiries to us: how many answered within ten working days | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Donations to the aid organisations, as confirmed by them | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Costs of the week: share covered | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
| Corrections to our texts | to be entered before 16 Nov | to be entered before 16 Nov | to be entered after 24 Nov | to be entered after 24 Nov |
"How confident" means: we state how likely we think it is that the expectation will be met, for example "7 in 10". Across many expectations you can then see whether we were too confident or too cautious.
What we do not claim
This way of working is not a tested method. We do not claim that it leads to better results; we do not know. It makes visible how far expectation and result lie apart, for us and for others. No outside body has reviewed or endorsed these commitments; they are our own.
The idea comes from working papers by Aciel Martinez Poll on expectation, result and feedback. They are unpublished and have not been validated. Related ideas have long existed elsewhere. Weather services compare their probability forecasts with what actually happens; a common measure for this was described by Glenn Brier in 1950. Philip Tetlock and others have collected forecasts on world affairs and scored them afterwards. In science, researchers use pre-registration to fix what they intend to test before they know the results. The evaluation criteria of the OECD Development Assistance Committee ask, among other things, to what extent a project achieved its objectives. We adopt the attitude behind these, not a claim to meet their standards.
Remind us
If we fail to keep one of these commitments, write to info@aciel-martinez-poll.com with the subject line "Standard". You will get a reply to that too, and every such message is counted on the feedback page.
This page describes a voluntary commitment by Helfen statt Blockieren. “Helfen statt Blockieren” is non-partisan. The German version of this website is legally binding. As of 3 November 2026.
Sources
- Glenn W. Brier, Verification of forecasts expressed in terms of probability, Monthly Weather Review 78(1), 1950, pp. 1 to 3, https://doi.org/10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2
- Philip E. Tetlock, Expert Political Judgment: How Good Is It? How Can We Know?, Princeton University Press 2005, https://search.worldcat.org/title/56825108
- Barbara Mellers et al., Psychological strategies for winning a geopolitical forecasting tournament, Psychological Science 25(5), 2014, https://doi.org/10.1177/0956797614524255
- Brian A. Nosek, Charles R. Ebersole, Alexander C. DeHaven, David T. Mellor, The preregistration revolution, PNAS 2018, https://doi.org/10.1073/pnas.1708274114
- OECD, Better Criteria for Better Evaluation: Revised Evaluation Criteria Definitions and Principles for Use, OECD DAC Network on Development Evaluation, December 2019, https://www.oecd.org/en/publications/better-criteria-for-better-evaluation_15a9c26b-en.html
- 22 U.S. Code § 6022, Purposes (Cuban Liberty and Democratic Solidarity (LIBERTAD) Act of 1996), Cornell LII, https://www.law.cornell.edu/uscode/text/22/6022
- Council Regulation (EC) No 2271/96 of 22 November 1996, EUR-Lex, https://eur-lex.europa.eu/eli/reg/1996/2271/oj
- Reply of the German federal government of 19 January 2026, Bundestag printed paper 21/3723, answer to question 7 (in German), https://dserver.bundestag.de/btd/21/037/2103723.pdf
All online sources accessed on 3 November 2026.