Before you even think about rewards or your campaign page, there is one choice that shapes your whole strategy: what happens if you don't reach your goal. There are two rules, “all or nothing” and “keep it all”. This guide explains how they work, who uses them, what the research says and how to decide.
Two rules, one question
All or nothing
You only receive the money if the campaign reaches its goal by the deadline. If it falls short, no backer is charged and the project receives nothing.
Keep it all
You keep what you raise even if you fall short of the goal. In return, the rewards you promised your backers must be delivered regardless.
There is really only one question to ask: with part of the money, can I deliver part of the project? If the answer is yes, “keep it all” protects you: you don't lose what you have raised. If the answer is no, “all or nothing” protects both you and your backers from a promise you cannot keep.
Who uses which
| Platform | All or nothing | Keep it all |
|---|---|---|
| Kickstarter | yes, exclusively | no |
| Indiegogo | yes, exclusively since 16 October 2025 | no longer for new campaigns |
| Ulule | yes, for reward campaigns | for donations and subscriptions |
| GoFundMe | no | yes (donations) |
| Campagna Crowdfunding | no | yes |
Sources: Kickstarter[5], Indiegogo[6], Ulule[7], GoFundMe[8].
There is a clear divide. The big international product platforms use “all or nothing”, because they are pre-selling items that can only be manufactured above a certain quantity. Donation platforms use “keep it all”, because for a cause every euro counts.
What the research says
The broadest comparison was carried out by Douglas Cumming, Gaël Leboeuf and Armin Schwienbacher on 22,850 Indiegogo campaigns between 2011 and 2013, when the platform offered both rules.[1]
| All or nothing | Keep it all | |
|---|---|---|
| Campaigns | 1,200 (5.2%) | 21,650 (94.8%) |
| Median goal | $16,485 | $10,000 |
| Campaigns reaching their goal | 34% | 17% |
| Median amount raised | $3,903 | $2,465 |
| Backers, median | 43 | 33 |
“All or nothing” reaches the goal more often, and the authors explain why: choosing it signals commitment, and those who choose it tend to propose larger, more structured projects. But their conclusion is not “always choose all or nothing”. It is that small, scalable projects are better suited to “keep it all”, while large, non-scalable ones suit “all or nothing”.[1] Two caveats are worth bearing in mind: this is a correlation, not a cause, and with “keep it all” even campaigns that fall short of the goal take home what they have raised.
Theoretical studies add some nuance. For the economist Jen-Wen Chang, “all or nothing” is the optimal choice when the problem is uncertainty about demand.[4] A Wharton School paper, still under review, finds that “keep it all” raises less on average, but can pay off more for projects with good sales prospects, because even a partial raise brings in cash and proof that people want the product, which helps in finding further funding.[2] And in an experiment on donations, donors coordinated better under “keep it all”: they spread themselves less thinly across too many projects.[3]
There is no best model. There is the right model for your project.
When keeping it all makes sense
- The project can be done in stages: a restoration in several phases, a festival with more or fewer nights, a record with more or fewer copies.
- The rewards scale with the amount raised: digital copies, experiences, thank-yous, print-on-demand products.
- It is a cause: for an association, a team or a school, every euro raised is a euro put to good use.
- You want to use the money straight away: on Campagna Crowdfunding the funds raised can be used during the campaign itself, month by month.
- You want a longer campaign, split into stages, rather than a thirty-day sprint.
How to set your goal with “keep it all”
With “keep it all” the goal doesn't decide whether you receive the money, but it still matters a great deal: it is the milestone you talk about, and your supporters become more active as the campaign gets close to its goal, especially when the goal is small.[10] Higher goals, in general, are reached less often.[9]
- Set the goal at the first stage that works on its own: the minimum amount with which the project genuinely exists.
- Add the following stages as new milestones, with new rewards, once the first is reached.
- Spell out on your page what you will do with each amount: with €3,000 this, with €6,000 that as well.
The promise still stands
We say it openly on our How it works page too: “keep it all” is the right model if part of the amount lets you deliver part of the project, and the wrong model if half the amount lets you do nothing, because the rewards must be delivered all the same. The answer is not to change model: it is to choose a goal and rewards you can honour whatever amount you raise. How to build them is explained in How to choose your rewards.
Frequently asked questions
What is the difference between “all or nothing” and “keep it all”?
With “all or nothing” you only receive the money if you reach your goal; otherwise nobody pays. With “keep it all” you keep whatever you raise, even below the goal, but you still have to deliver the rewards you promised.
Which model is more successful?
In the only large study on the subject, covering 22,850 Indiegogo campaigns from 2011-2013, 34% of “all or nothing” campaigns reached their goal, compared with 17% of “keep it all” campaigns. But with “keep it all”, even campaigns that fall short of the goal keep what they have raised, and the authors conclude that it is the better fit for small, scalable projects.
Which projects suit “keep it all”?
Projects that can be carried out in part: a record with fewer copies, a smaller event, a restoration in several phases, a cause where every euro counts. It is risky, on the other hand, if half the amount lets you achieve nothing at all while you still have to deliver the rewards.
Which platforms use “keep it all”?
GoFundMe for donations and Campagna Crowdfunding for rewards. Kickstarter and, since 16 October 2025, Indiegogo use only “all or nothing”; Ulule uses “all or nothing” for reward campaigns.
Sources
- Financial Management, Cumming D.J., Leboeuf G., Schwienbacher A., Crowdfunding models: Keep-It-All vs. All-Or-Nothing, 49(2), pp. 331-360 (data from the working paper version of 31 May 2015), 2020. ideas.repec.org/a/bla/finmgt/v49y2020i2p331-360.html
- Wharton School (working paper), Astashkina E., Marinesi S., All-or-Nothing vs Keep-It-All: Comparing Campaign Designs in Rewards-Based Crowdfunding Platforms, 2023. faculty.wharton.upenn.edu/wp-content/uploads/2016/11/Choice_of_Design_in_Crowdfunding_AoN_vs_KiA_24.08.2023.pdf
- ACM CSCW, Wash R., Solomon J., Coordinating Donors on Crowdfunding Websites, pp. 38-48, 2014. rickwash.com/papers/conference/return-rule.html
- American Economic Journal: Microeconomics, Chang J.-W., The Economics of Crowdfunding, 12(2), pp. 257-80, 2020. aeaweb.org/articles?id=10.1257/mic.20170183
- Kickstarter Help Center, What is Kickstarter?.
- Indiegogo Help Center, What changed with the platform upgrade.
- Ulule, Condizioni generali [General terms and conditions].
- GoFundMe, Domande frequenti [Frequently asked questions].
- Journal of Business Venturing, Mollick E., The dynamics of crowdfunding: An exploratory study, 29(1), pp. 1-16, 2014. papers.ssrn.com/abstract=2088298
- Journal of Business Venturing, Kuppuswamy V., Bayus B.L., Does my contribution to your crowdfunding project matter?, 32(1), pp. 72-89, 2017. doi.org/10.1016/j.jbusvent.2016.10.004
