Maria, coordinator of the civic budget in the booming city, looks proudly at the turnout charts. 30% more residents participated in this year’s edition than in the previous one. Digitization of the process and an intuitive voting app have yielded results. But as she delves into the detailed results, her enthusiasm wanes. Once again, the same frustrating pattern repeats itself. Two large, media-savvy projects - the construction of an aquapark and the modernization of a stadium - garnered more than 80% of the total budget. They won because they were supported by well-organized interest groups that mobilized their supporters. As a result, dozens of smaller but extremely valuable initiatives submitted by local communities - building a playground, creating a pocket park, installing lighting on a neighborhood street - received no funding at all. The “winner takes all” voting system has led to polarization and disappointment. On social media, Maria reads comments full of bitterness: “Why vote if the same, biggest projects always win anyway?”. She begins to wonder: are we sure we’ve chosen the best projects for the whole city, or just the ones with the best promotional campaign? Is there a way to make the voting process fairer, smarter and better reflect the diverse needs of all residents?

Maria’s dilemma is one faced by organizers of participatory budgets around the world. Citizen budgeting is one of the most beautiful ideas of direct democracy, but its success depends hugely on the “mechanics” - on how the process of collecting ideas and, most importantly, voting is designed. The simple methods we know from political elections, in the context of allocating scarce public resources, often lead to suboptimized and unfair results. Fortunately, thanks to a combination of social choice theory, advanced mathematics and modern technology, we are entering a new era of participation. An era in which we can design voting systems that are not only easy to use, but also smart. This article is a journey into the future of citizen budgeting. We will show how innovative voting methods, artificial intelligence and other breakthrough technologies can help us make better, collective decisions about the future of our cities and communities.

Why is the traditional “one vote per project” model inadequate for a modern civic budget?

“65% of respondents report that their organizations are regularly using generative AI, nearly double the percentage from ten months earlier.”

— McKinsey & Company, The State of AI in Early 2024 | Source

The voting model, in which each resident has one vote and can cast it for one project of his or her choice, is seemingly the simplest and most intuitive solution. This is classic majority voting (plurality voting). However, this simplicity is its biggest drawback. In the context of a civic budget, where we are dealing with a wide variety of projects and a limited budget, this model leads to a number of pathologies and unintended negative consequences.

1 The problem of “winner takes all” and wasted votes: Under this system, the projects that win the most votes, often slightly ahead of others, win. This leads to a situation in which a project that has won the support of, for example, 20% of the voters can claim a significant share of the budget, while the votes of the remaining 80% of residents, cast for other, also popular projects, are in practice “wasted.” They have no impact on the final result. This naturally leads to frustration and a feeling that there is no point in participating in voting.

2. susceptibility to polarization and strategic voting: the system promotes “big” and “controversial” projects that can mobilize large but often narrow groups of supporters. Moderate, compromise projects, which enjoy broad but less fervent support, have little chance of winning. Moreover, this leads to so-called “tactical voting.” A resident, fearing that his favorite niche project won’t win anyway, may “tactically” vote for another project that is more popular but less to his liking, just to block the win of the project he dislikes most. This distorts voters’ real preferences.

3. ignoring the cost of projects: In the simplest form of this model, each vote carries the same weight, regardless of whether it is cast for a £5,000 bench project or a £5,000,000 sports hall project. This leads to inefficient budget allocation. Often, one very expensive project, which won by a small margin of votes, “eats up” the entire budget, preventing the implementation of dozens of smaller, cheaper and perhaps more necessary projects, which together would benefit more residents.

4 Limited expression of preferences: By forcing a resident to choose only one project, the system does not allow him to express the full spectrum of his preferences. A resident can support five different projects to varying degrees, but the system forces him to make a zero-one decision. We learn nothing about which other projects he considers valuable.

These flaws do not mean that the idea of a citizen budget is bad. They just mean that we need smarter, more sophisticated decision-making tools. Tools that allow us to better reflect the collective will of the community.

What is approval voting and how does it promote compromise projects?

Approval Voting is a simple but extremely powerful alternative to traditional voting that elegantly solves many of its problems. The principle is trivially simple: instead of choosing just one project, a voter can “approve” (vote for) any number of projects he or she supports.

There is no ranking or awarding of points. Each project that a resident deems valuable receives one vote-approval from him. Once the voting is over, projects are ranked according to the number of approvals they receive, and funding goes to those with the highest support until the budget is exhausted.

What problems does this method solve?

1 It eliminates the problem of “wasted votes.” Since you can vote for all the projects you like, you don’t have to worry that your “main” candidate won’t win. You can safely support both your favorite niche project and several other, more popular ones that you also think are good for the city. Every vote you make matters.

2. promotes projects with broad support (compromise projects): This is the biggest advantage of this method. In traditional voting, the project with the most ardent supporters wins. In an approval vote, projects that are acceptable to the broadest group of residents win. A controversial project that is loved by 25% of residents, but disapproved of by the other 75%, will lose out to a project that may not be anyone’s “first choice”, but is viewed favorably by 60% of the community. This promotes consensus and the selection of projects that benefit the entire city, not just one group.

3 It reduces negative campaigning and polarization: In the traditional system, in order to win, one often has to not only promote one’s project, but also discourage voting for the competition. In approval voting, such a strategy makes less sense. Support for one project does not come at the expense of another. It encourages a more substantive and positive discussion of the merits of each proposal.

4 It is simple to understand and implement: Despite its mathematical elegance, this method is extremely simple for voters to understand. The instruction to “mark all the projects you support” is clear and intuitive. From a technological perspective, implementing such a voting system is also relatively simple and does not require significantly more complexity than traditional voting.

Approval voting is a powerful tool that allows for much better and fairer aggregation of community preferences. It is one of the first and most important steps toward building smarter participatory systems.

What is the “knapsack budget” (knapsack voting) method and how does it allow for smarter allocation of resources?

If approval voting is a step toward a better reflection of preferences, then the “knapsack budget” (Knapsack-Based Voting) method is a revolution in the way we think about allocating scarce resources. The method draws its name and logic from a classic computer science problem known as ** the “knapsack problem.”**

Imagine you are out shopping and you have a backpack with a limited capacity (e.g., 10 kg) and a number of items, each of which has its own weight and value. Your goal is to pack a set of items in your backpack that will fit within the weight limit while having the highest total value possible.

How does this translate into a civic budget?

In the “backpack budget” method, each resident is given a virtual “backpack” whose capacity is the total amount of the civic budget. His task is to “pack” into this backpack such a set of projects that he considers best for the city, provided that the total cost of these projects does not exceed the budget limit.

Once all residents have “packed” their ideal backpacks (i.e. voted for their chosen sets of projects), the system moves on to the results aggregation phase. The goal of the algorithm is to find the set of projects that is closest to the “collective will,” i.e. is most similar to the proposals of as many residents as possible, and at the same time fits within the real budget of the city. This is a complex optimization problem, but modern computer systems can solve it effectively.

The main advantages of the backpacking method:

1 It takes into account the cost of projects: This is a fundamental change. A voter no longer casts a vote for a single project in isolation from its cost. He has to make informed decisions and compromises. Do I prefer one large, expensive project, or ten smaller, cheaper ones? This forces you to think in terms of a realistic budget and the value that every zloty spent brings.

2. allows the expression of complex preferences: A resident can express much more than just “I support” or “I don’t support.” He or she can create a whole, coherent “package” of investments that he or she believes will best serve the city. The system is able to aggregate these composite preferences.

3 It leads to more optimal results: The algorithm, looking for the “best fit,” is able to find solutions that maximize the “satisfaction” of as many residents as possible. Instead of a single, polarizing project, the system can select a set of several smaller ones that were part of the “backpacks” of many different voting groups, leading to a fairer and more efficient distribution of resources.

4. educates residents: The “backpack packing” process also has great educational value. It makes residents aware that the budget is limited and that every spending decision involves an opportunity cost - by choosing one thing, we give up something else. It’s a lesson in responsibility and micro-management of public finances.

The backpack method is much more advanced than simple voting, but with modern technology platforms such as ARDVote from ARDURA Consulting, its implementation is becoming possible. This is a step toward participation that is not only democratic, but also smart and optimized.

What role can blockchain technology play in ensuring transparency and trust in e-voting?

In any voting process, and especially in electronic voting, trust is absolutely fundamental. Participants must have confidence that the system is secure, that their vote has been recorded correctly, that no one can change it, and that the vote counting process is fully transparent and impossible to manipulate. Traditional, centralized database systems, while they can be very secure, always carry the risk (or perceived risk) of manipulation by the system administrator. **Blockchai ** technology offers a radically new approach to building trust.

What is blockchain? Blockchain is a decentralized, distributed and immutable (immutable) database. Instead of being stored on one central server, it is replicated and shared by many computers on a network. Each new transaction (in our case - a vote cast) is grouped into a “block”, which is cryptographically “linked” to the previous block, forming an inseparable “chain”.

How can blockchain increase trust in e-voting?

1 Immutability (Immutability): Once a voice is stored in a block and added to a chain, it is virtually impossible to remove or change. Any attempt to modify one block would require changing all subsequent blocks in the entire chain, which is cryptographically unfeasible. This gives the voter a guarantee that his vote will not be changed once cast.

2 Transparency and auditability: Since blockchain is inherently distributed, a copy can be publicly available (in anonymized form). Any interested party (e.g., watchdog organizations, journalists, or even residents themselves) can independently verify and count the votes, confident that no one has manipulated the results. The counting process becomes fully transparent.

3 Decentralization and resilience to attacks: In a traditional system, an attack on a central server can destroy or block the entire voting process. In a blockchain-based system, there is no single, central point of failure. An attack on one network node does not affect the operation of the others.

4 Verifiability for the voter: Advanced blockchain-based voting systems can provide each voter with a unique, anonymous “proof” of having cast a vote. With this proof, a voter can verify with the public record (blockchain) that his or her vote is indeed included in the pool, without revealing his or her identity or what he or she voted for.

Challenges and limitations: However, it is important to remember that blockchain is not a magic solution to all problems. Implementing it in voting systems comes with challenges, such as:

  • **Anonymity vs. authorization: ** How to ensure that only eligible people can vote (and only once), while guaranteeing full anonymity of the vote itself?

  • Scalability: Public blockchain networks (like Bitcoin or Ethereum) can be slow and expensive. Voting systems require more efficient, often private or consortial, blockchain implementations.

  • Complexity and user experience: The system must be designed so that all technological complexity is hidden from the end user, for whom the process must remain simple and intuitive.

Despite these challenges, the potential of blockchain to build a new generation of transparent and trustworthy democratic systems is enormous. It is one of the most promising developments in the Civic Tech world.

How can artificial intelligence (AI) and chatbots increase engagement and educate residents?

One of the biggest challenges in civic budgeting is the low turnout and information barrier. Residents often don’t know the budget exists, don’t understand its rules, or feel overwhelmed by the number and complexity of submitted projects. Artificial intelligence, particularly conversational AI (chatbots) and generative AI, offers powerful tools to break down these barriers, personalize communication and actively engage communities.

1. intelligent participation assistant (Chatbot): Imagine a chatbot integrated with the city’s website or messenger (e.g. Messenger, WhatsApp). Such an assistant, available 24/7, could:

  • Answer the questions: “Who can vote?”, “Until when can projects be submitted?”, “What are the rules?”.

  • To help navigate: “Show me all the sports projects in my district”, “Find bike path projects that cost less than £100,000”.

  • Guide through the voting process: Step by step help the user to cast a vote, explaining the rules of the chosen method (e.g. backpack budget).

  • Send personalized reminders: “3 days left to vote! Do you want to cast your vote?”.

2 Generative AI as support for project developers: Submitting a good, well-described project is difficult, especially for those without experience. Generative AI (such as GPT models) can act as a “creative partner.”

  • Proposal writing assistance: The project proponent describes his idea in a few simple sentences, and AI helps him dress it up in a formal, compelling and compliant project description.

  • Preliminary cost estimation: Based on the project description and cost data of similar past implementations, AI can help prepare a preliminary, indicative cost estimate.

  • Identification of potential problems: AI can analyze the description for compliance with the budget regulations and identify potential problems (e.g., “This project appears to be located on private property, make sure you have the owner’s permission”).

3 Personalization and recommendations: By analyzing demographic data (anonymized and aggregated, of course) and historical voting patterns, an AI-based system could:

  • Recommend projects: Similar to the way Netflix recommends movies, the voting platform could suggest projects of interest to residents based on their location, interests or previous votes.

  • Provide personalized information: Instead of sending all residents the same general information, the system could send notifications tailored to their profile (e.g., “A new playground project has been submitted in your neighborhood!”).

4 Sentiment and public opinion analysis: AI can analyze project discussions on social media and online forums to give budget organizers valuable insight into public sentiment, major points of contention and arguments raised by supporters and opponents of particular initiatives.

Using AI in participation is not about replacing a human decision with a machine decision. It’s using technology to make the democratic process more accessible, understandable and engaging for every resident.

What technological challenges do modern e-voting systems face?

Creating a modern, secure and reliable electronic voting platform is one of the most demanding tasks in the world of software engineering. These systems have to meet a unique set of non-functional requirements, where the stakes are extremely high - public trust in the democratic process.

**1 Security - an absolute priority: A ** e-voting system is a top priority target for potential attackers. Security must be built into every layer of the system, from infrastructure to application code.

  • Protection against DDoS attacks: the platform must be able to withstand massive Distributed Denial of Service attacks that aim to block access to the service at key voting moments.

  • Integrity and confidentiality of the vote: Advanced cryptographic techniques should be used to ensure that a vote, once cast, caot be altered (integrity) and that no one is able to link a particular vote to a particular person (anonymity).

  • Application security: Application code must undergo rigorous security testing (SAST, DAST, penetration testing) in accordance with DevSecOps best practices to eliminate vulnerabilities such as SQL Injection and Cross-Site Scripting.

2 Scalability and performance: Citizen budget voting has a very uneven load. Most of the time there is little traffic on the platform, but in the final hours before voting closes, there can be a huge surge of users.

  • Flexible scalability: The system architecture must be cloud-based and designed to scale up automatically and instantly to handle peak load, and then scale down to optimize costs. ARDVote, ARDURA Consulting’s platform, is designed to handle a load of 1,000 voices per minute, ensuring fluidity even in the largest cities.

  • Performance under load: The system must be subjected to advanced performance tests that simulate real-world scenarios and ensure that the application remains responsive even at the peak of interest.

3 Digital Accessibility (Accessibility): A public system must be accessible to everyone, without exception. This includes people with disabilities (e.g., blind, visually impaired, mobility impaired), the elderly, or people with lower digital skills.

  • WCAG compliance: The platform must be designed and coded in accordance with the international Web Content Accessibility Guidelines (WCAG), at the level required by law (now usually 2.1 or 2.2, as in the case of ARDVote). This means, among other things, full keyboard capability, compatibility with screen readers, adequate color contrast and understandable structure.

  • Simple and intuitive interface (UX): The interface design must be as simple, clear and intuitive as possible to minimize the barrier to entry for all users.

4 Auditable and transparent: As mentioned in the context of blockchain, the entire system must be fully auditable. Every event should be recorded (in a secure and anonymized ma

er) so that in case of any doubts or disputes, the entire process can be accurately reconstructed and verified.

Building such a platform requires not only excellent programming skills, but also deep knowledge of cyber security, reliability engineering (SRE) and universal design.

What are the key criteria for selecting a next-generation citizen budgeting platform?

Choosing the right technology platform is one of the most important decisions that civic budget organizers make. It’s a decision that affects not only the efficiency of the process, but also the level of citizen engagement and their confidence in the overall initiative. When choosing a solution for the future, it is important to go beyond basic functionality and evaluate the platform in terms of its strategic capabilities.

1 Flexibility and support for advanced voting methods: The platform should not impose a single, rigid voting model. It should be a flexible tool that allows for easy configuration and experimentation with a variety of modern methods, such as approval voting or the backpack method. This is crucial for being able to adapt the process to the specifics and goals of the community.

2 Safety and compliance proven by certifications: Safety caot be just a marketing promise. Hard evidence should be required from the vendor. Does the platform undergo regular independent security audits and penetration testing? Does the vendor have certifications proving its information security competence (e.g., ISO 27001)?

3. documented scalability and reliability: the vendor should be able to provide performance test results and document what kind of load its platform can handle. Ask for case studies (case studies) from deployments in cities of similar or larger scale. How did the platform perform during peak hours? What was its availability rate (SLA)?

4 Digital accessibility (WCAG) and excellent user experience (UX): Compliance with the WCAG standard at the required level should be an absolute prerequisite. Ask for an accessibility declaration and, if possible, the results of an accessibility audit conducted by an external company. The interface should be simple, modern and intuitive, both for residents and administrators.

5 Comprehensive support for the entire process: A modern platform is more than just a voting module. It should support the entire life cycle of the civic budget: from the secure and intuitive submission of projects by residents, through the stage of formal and substantive verification by officials, to the voting itself and transparent presentation of the results.

6 Technology partner experience and support: You’re not just choosing software, but a partner. Does the provider have real experience in implementing citizen budgets? Does it offer professional technical and substantive support at every stage of the process? Is it a partner that can advise and help solve unusual problems?

Choosing a platform is a strategic investment in the quality of local democracy. It is worth taking the time to carefully analyze and choose a solution that is not only functional today, but also ready for the challenges and opportunities of tomorrow.

Need testing support? Check our Quality Assurance services.

Let’s discuss your project

Have questions or need support? Contact us – our experts are happy to help.

How does the ARDVote platform from ARDURA Consulting address the challenges of the future of public participation?

At ARDURA Consulting, we believe that technology can be a powerful force for a more engaging, transparent and intelligent democracy. Our proprietary platform, ARDVote, is the product of this belief. It is not just another voting tool. It’s a comprehensive, modern and secure solution, designed from the ground up with the challenges and opportunities of the future of public participation in mind.

Flexibility and readiness for innovative voting methods: We understand that the future of citizen budgets lies in smarter decision-making methods. ARDVote is designed in a modular and flexible way to support not only traditional voting, but also advanced models such as approval voting and the innovative backpack method, giving organizers the tools to build fairer and more optimized processes.

Uncompromised security and scalability: Trust is the foundation of democracy, so security is in the DNA of our platform. ARDVote is regularly subjected to rigorous security testing. Its architecture, based on modern DevSecOps practices, is designed to be resilient to attacks. At the same time, thanks to its flexible, cloud-based architecture, the system is able to handle the largest workloads, guaranteeing performance of more than 1,000 votes per minute, ensuring a smooth process even in the largest metropolises.

Accessibility and inclusivity as a priority: participation is for everyone. That’s why ARDVote has been designed and implemented in accordance with the highest standards of digital accessibility, meeting the requirements of WCAG 2.2. We make sure that every person, regardless of age, ability or digital competence, can easily and intuitively participate in the decision-making process.

Comprehensive support for the entire process: ARDVote is an integrated ecosystem that supports the entire citizen budget lifecycle - from the moment a resident has an idea to the a

ouncement of the results. Our platform includes modules for:

  • Submit projects with attachments and location on a map.

  • Formal and substantive **verification ** by dedicated teams of officials.

  • Secure and intuitive online voting.

  • Transparent presentation of results through interactive dashboards and maps.

ARDURA Consulting partnership and experience: when you choose ARDVote, you not only get software, but also a partner with years of experience in developing and implementing complex IT systems. The ARDURA Consulting team provides full support at every stage - from configuration and implementation to training and ongoing technical support.

ARDVote is our answer to the future of participation. It’s a platform that combines the best of technology - security, scalability and innovation - with the best of democracy - inclusivity, transparency and collective wisdom.

If you’re looking for a partner to help you take civic budgeting in your community to the highest, world-class level, consult your project with us. Together we can build the future of engaged cities.