Three years ago, we started a bold journey: to help companies win public contracts through AI-driven tender monitoring, analysis, and real-time market intelligence.

From idea to impact

Hermix was created by IT people with 60+ years’ of total experience in IT and public procurement. We know the market, and we understand its needs.

The Business-to-government (B2G) sector is great, and it worked very well for us: stable market, lots of money, and lots of information, if you know where to look and how to read.

3 years changed Hermix a lot, and this is what we, the people behind this platform, feel about it:

We are deeply grateful to our clients and users

Our newest clients are Kyndryl, Serco, NRB (Trasys), Deep – Post Telecom, QUALCO, Fujitsu, SopraSteria, Unisys, Intrasoft, Cancom, and Nova ICT. See all clients

We are proud to be part of Microsoft for Startups, Amazon AWS Activate, the NVIDIA Inception program, and Eureka Eurostars, and to partner with prestigious institutions such as the University of Luxembourg, the University of Bucharest, and KU Leuven. Through these collaborations, we achieved a significant milestone: winning a €920K grant from Eureka Eurostars in partnership with the University of Luxembourg.

Our client’s success is our success

• Our clients Capgemini and Unisystems won a €6.6 million contract for IT Operations, Engineering, and Evolution services.
• Hermix’s clients, ARHS, Aricoma (Qinshift), Fincons, Thaleria, Unisys, Westpole, and VASS are among the companies awarded the contract for “Provision of external service providers for DG TAXUD.”
• Eight of the companies awarded the contract for “IT services: consulting, software development, Internet, and support European Parliament”,  Aricoma, Cleverlance Autocont, Engineering Group, Fincons, Qualco, Seidor, and Trasys – Network Research Belgium, are Hermix’s clients.

These used Hermix to help them find and win this public tender. With Hermix’s AI-powered tools, they could track opportunities, understand the market, and make better decisions in real time. This helped them stand out and succeed in a competitive process.

Clients love is everything 

“Hermix makes our work easier, especially when analyzing past years of public data across clients, providers, or subjects. It’s clear, visual, and gives us a strong starting point for AI-driven analysis and tender preparation,”  said Arian Turhani, Vice President and Client partner for European Institutions at Capgemini

”An amazing, powerful platform that stands as a solid improvement over all other tools for public sector work,” said Erik Francq, Account Director European Institutions at Fujitsu

Every feature & improvement changes everything

Over the past three years, we’ve launched powerful features such as AI Tender summarization, AI Chat, AI-powered Q&A classification, and added new data sources for improved efficiency and deeper market insights.

• AI Tender Summarization: Summarize thousands of public tender documents with AI. You can summarize tender documents from multiple countries, including Belgium, Norway, Romania, and the Netherlands, e-licitatie, EU Funding portal with just one click.
• AI Chat: Customize your questions and obtain personal insights from a specific tender opportunity in just a few seconds. It helps you navigate complex documents effortlessly. It supplements the AI tender summarization.
• AI-powered Q&A classification: AI organizes and classifies Q&As to make key questions and answers easier to find and read. You can export the results as PDF, Excel, or HTML files.
• We added new data sources. We now monitor tenders and clean-up data from TED (EU-wide tenders, European and international institutions), e-tendering, the EU funding portal, various European institutions websites, NATO, Eurostat, ECB, as well as national portals from Belgium, the Netherlands, Norway, Romania, Germany, UK, NATO, Spain, France, and more. See all features

Competition fuels our growth

We received international recognition. We won 1st place at the EU Datathon 2022 competition, which took place in Brussels in October 2022. 

We are also happy that our work was awarded three times in a row by Deloitte & Google: Deloitte’s Impact Star award in November 2022 & Deloitte’s CE Tech Rocketship in November 2023 and 2024.

Collaboration is key

Year after year, we value our team and meetings more and more, as they help us better understand each other’s perspectives and contribute together to Hermix’s innovation.

To many more years of Hermix!

Sign up now  to Test Hermix for free: https://hermix.com/sign-up/

The integration of artificial intelligence (AI) and advanced data analytics into public sector sales and procurement is reshaping how companies interact with complex tendering processes. This transformation is not just about automation but about leveraging smart technology to gain competitive advantages in a market traditionally marked by intricate regulations, voluminous documentation, and high-stakes decision-making. 

Beyond automation and efficiency, a growing debate is emerging: Is generative AI the beginning of the end or the dawn of a new era in the way we work? While some envision a utopia of effortless productivity, others fear a dystopian world where AI replaces human judgment. For public sector professionals, the truth likely lies in between navigating risks while embracing extraordinary opportunities.

In this comprehensive article, we take a closer look at a recent live webinar hosted by Hermix, a leader in public sector sales analytics, with Rudolf de Schipper, General Manager at Unisys Belgium, and Stefan Morcov, CEO of Hermix. Together, they discuss how AI-driven tools are revolutionizing tender analysis, bid automation, market intelligence, and proposal writing in public procurement.

Drawing on decades of combined expertise in public procurement, software development, project management, and sales, Stefan and Rudolf provide practical insights into the current capabilities, challenges, and future directions of AI in this specialized field. Whether you are a business leader, a sales professional, or a public procurement specialist, this article will help you understand how to harness AI for smarter decision-making and enhanced efficiency in public sector sales.

Table of Contents

1. Understanding the Impact of AI in Public Sector Procurement and Sales
2. Data-Driven Analytics: Big Data and Machine Learning in Tender Analysis
3. GPT and Large Language Models for Tender Document Analysis and Proposal Writing
4. Practical Applications and Real-World Examples from Hermix
5. Challenges and Limitations in AI Adoption
6. Future Trends and Ethical Considerations in AI for Public Procurement
7. Frequently Asked Questions
8. Conclusion

About speakers

Rudolf de Schipper

General Manager at Unisys Belgium has extensive experience managing European and international public sector business.

Stefan Morcov

CEO of Hermix, with over 20 years of experience in complex IT projects for EU institutions across Europe.

Understanding the Impact of AI in Public Sector Procurement and Sales

Artificial intelligence is no longer a futuristic concept limited to research labs or science fiction. It is actively transforming industries worldwide, and public sector procurement is no exception. AI and related technologies are reshaping how companies identify tender opportunities, analyze market data, qualify bids, and even write proposals.

This transformative power of AI can be compared to historic technological shifts such as the replacement of horses by the Ford Model T or the evolution from paper-based communication to instant digital messaging. Just like Excel revolutionized spreadsheets, or the iPhone changed how we work on the go, AI is enhancing how sales teams handle complexity and volume.

While AI is gaining attention, it is still challenging to predict its long-term impact. The technology is evolving fast, and the industry is in a phase of exploration, balancing enthusiasm with caution. The human element continues to play a crucial role, particularly in framing the right questions for AI systems and validating their outputs.

The Dual Role of AI: Automation and Augmentation

A key theme is the dual role of AI in public sector sales: automating repetitive tasks and enhancing human decision-making. AI can automate tender monitoring, data cleanup, and initial qualification, saving valuable time for sales teams to focus on strategy and relationship-building. At the same time, AI augments human capabilities by providing deep insights into market trends, competitor behavior, and partner networks.

Public sector sales involve answering two fundamental questions:

• Do I want to pursue this tender? (Qualification)
• How should I approach this tender? (Strategy and execution)

AI tools assist in answering these questions by analyzing vast datasets, summarizing complex tender documents, and providing competitive intelligence. This blend of automation and augmentation is critical to navigating the complexity and volume of public procurement opportunities.

Utopias, Dystopias, and Practical Realities

While the benefits of AI are clear, so are the concerns. Two opposing narratives dominate public discussion: the utopian view of AI as a liberator from boring work, and the dystopian fear of AI as a savior or destroyer, like Messiah vs Skynet; Terminator or Matrix. Technologically plausible? Yes. Desirable or fair? Not necessarily. This underlines the importance of human oversight and ethical frameworks.

Imagine a world where:

• Authorities write RFPs with AI
• Companies write proposals with AI
• Evaluations are done by AI
• Winners are picked based on price alone

In public procurement, the practical reality is somewhere in the middle. AI helps us do a significantly smarter work, more efficiently, and increase our win-rate.

Data-Driven Analytics: Big Data and Machine Learning in Tender Analysis

At the core of the AI revolution in public sector sales is the use of big data and machine learning to analyze procurement markets. With Hermix providing access to a clean dataset of over eight million public contracts across Europe, along with detailed company profiles and authority records, it becomes possible to perform advanced analyses that were once either impossible or far too time-consuming.

Key Data Analytics Capabilities

Some of the core data-driven analytics capabilities include:

• Smart Tender Matching: Identifying tenders that closely align with a company’s capabilities and strategic goals.
• Market Intelligence: Assessing the size, value, and trends within specific markets or sectors, such as IT contracts in Luxembourg.
• Competition Analysis: Profiling competitors by analyzing their past contract awards, partnership histories, and geographic footprints.
• Partner Identification: Discovering and evaluating potential consortium partners based on historical collaboration data.
• Price Analysis: Evaluating pricing trends and benchmarking offers, although limited by confidentiality constraints on financial data.

Graph Analysis and Network Mapping

Graph analysis is an emerging technique used to model relationships between clients, suppliers, and contracts in public procurement. By representing entities as nodes and contracts as links, Hermix is able to detect subgraphs, identify frequent partnerships, and uncover anomalies. This approach provides a visual and quantitative understanding of market ecosystems, helping companies optimize their consortium strategies and competitive positioning.

Although the underlying data is publicly available, the true value lies in how it is cleaned, structured, and analyzed to generate actionable insights. Additionally, a new discipline is gaining importance: prompt engineering. Crafting precise questions for AI systems is becoming a key skill for extracting relevant and meaningful information from complex datasets.

GPT and Large Language Models for Tender Document Analysis and Proposal Writing

One of the most exciting developments in AI for public sector sales is the application of large language models (LLMs) like GPT for analyzing tender documents and assisting with proposal writing. These AI models can process complex, multi-page tender documents to generate concise summaries, answer specific questions, and even produce draft proposal content.

Automated Tender Summarization and Chat

Hermix includes an AI-powered tender summarization feature that extracts key information such as objectives, budget, award criteria, deliverables, and eligibility requirements. This helps sales teams quickly assess whether a tender is worth pursuing, without the need to manually read through hundreds of pages.

Additionally, Hermix offers an AI tender chat function that allows users to ask natural language questions about specific tender requirements. For example, users can inquire about expert qualifications, work package details, or financial thresholds and receive instant, accurate responses. This interactive tool enhances understanding and speeds up decision-making.

Proposal Writing and Bid Automation

AI has the potential to assist in writing proposals by generating coherent and well-structured text. However, AI-generated content should not be blindly trusted. The technology excels at compiling existing knowledge but lacks true creativity and contextual understanding.

AI-generated tender abstracts can be compared to book summaries, helpful for a quick overview, but not sufficient for deep analysis. Final proposal documents still require human review, interpretation, and original input to ensure quality, compliance, and clear differentiation.

Practical Applications and Real-World Examples from Hermix

Before diving into specific use cases, it’s important to understand what Hermix brings to the table.  Built for public sector sales and bid teams, Hermix combines AI and data analytics to simplify the complexities of public procurement. It helps teams track tenders across Europe, analyze markets, understand competitors, and make informed decisions quickly and confidently. 

The platform is already used by leading companies across multiple countries and industries, like Fujitsu, Capgemini, Indra, Atos-Eviden, Publicis-Sapient,  Intellera (formerly PwC, currently Accenture), Unisys, Technopolis, Kyndryl, Mindit, and more, helping them win more by working smarter.

Hermix has clients from Belgium, Luxembourg, Italy, Germany, UK, Spain, Greece, Portugal, France, US, India, Netherlands, Sweden, Romania, Poland, Czechia.

It provides clear examples of how AI and data analytics help public sector sales professionals in their daily work. These include:

Strategic Planning & Market Intelligence

This typically involves defining annual market strategies, building long-term pipelines, and analyzing large datasets of procurement activity. This task is done yearly and updated quarterly, in slides, spreadsheets, and diagramming applications. The market research requires 2-3 weeks of effort per year. The process relies on heavy data processing, analysis, and interpretation, suitable for data engineering and data science tools, especially based on AI/ML.

Tender Monitoring and Qualification

Hermix continuously downloads, indexes, and analyzes tender notices from multiple European portals, including TED (Tenders Electronic Daily), EU funding portals, e-tendering, EC, Eurostat, ECB, national portals: Belgium, Netherlands, Norway, Romania, Spain, France, Luxembourg, Germany, UK (Gov.UK – Contracts finder, Find a tender; Scottish Government; Sell2Wales), and NATO public procurement. Users receive alerts for new tenders that match their criteria and can quickly qualify opportunities based on AI-generated summaries and commercial intelligence.

Market and Competition Analysis

Using cleaned and comprehensive datasets, Hermix visualizes market trends, contract volumes, and values by country and sector. The platform also profiles competitors, showing their contract histories, geographic presence, and partnership networks. This enables users to identify “blue ocean” opportunities with high budgets and low competition.

Commercial qualification of tenders

This is performed daily and leads to a bid or no-bid decision. The process usually takes about 30 minutes but involves 4-5 days of manual effort per tender. It relies on analyzing large amounts of historical data, as well as client and company profiles. Smart tools are needed for intelligent technical and commercial analysis. New technologies like AI/ML, graph analysis, and anomaly detection are very useful for handling this kind of data processing.

Consortium and Partnership Insights

Hermix displays detailed partnership histories, allowing companies to evaluate potential consortium members. Users can view the number and value of contracts won together, the nature of collaborations, and graphical representations of partnership networks. This information helps in forming winning consortia tailored to specific tenders.

Price Analysis

Price analysis is based on historical data about budgets, competition, and pricing, along with a 100% financial evaluation using Taxud/Timea. 

Hermix already covers tender commercial context, client and company profiles, and market analysis. It uses analytics, data cleanup, graph analysis, predictive analysis, anomaly detection, machine learning, and GPT. There is a lot of data, and it’s not easy to find. Hermix shows where the money is, who is buying and selling, when, and how.

Technical qualification

It requires reading and understanding hundreds or even thousands of pages of specifications and technical annexes. This process can take anywhere from 30 minutes to 3 or 4 days per tender. However, the task can be significantly simplified with AI, especially with tools like LLMs and GPT.

Proposal automation and generation

This is a heavy task that involves preparing hundreds or even thousands of forms, such as CVs and references, along with detailed technical proposals, budgets, and project plans. For each tender, this effort can add up to hundreds or thousands of workdays. AI tools like GPT are extremely useful for automating forms, checking style, spelling, and formatting, and generating summaries, images, and diagrams such as Gantt charts, organizational charts, company profiles, and expert lists. 

However, current GPT tools do not generate original ideas and may miss contextual information, so human supervision remains critical. Hermix already provides tender summaries and Tender Chat using LLM/GPT tools. The approach is pragmatic: automate the critical repetitive tasks.

Challenges and Limitations in AI Adoption

AI brings important advantages, but it also comes with clear limitations and challenges, especially for public sector sales professionals:

Lack of Trust and Validation

One of the biggest barriers to AI adoption is skepticism about the accuracy and reliability of AI outputs.  For example, GitHub Copilot generated incorrect code 40% of the time, highlighting the need for human validation. In public procurement, where errors can jeopardize multi-million-euro contracts, trust in AI is paramount.

Contextual Understanding and Creativity

AI struggles with complex contextual nuances that are often critical in tender analysis and proposal writing. While AI can summarize requirements, understanding broader strategic implications or crafting innovative solutions remains a human domain. AI is a tool to assist, not replace, human creativity and judgment.

Data Confidentiality and Completeness

Public procurement data is extensive but incomplete, especially regarding unsuccessful bids and financial proposals. This limits AI’s ability to provide a full picture, particularly in price analysis and competitive intelligence.

Fragmented AI Ecosystem

AI landscape consists of multiple specialized tools rather than a single comprehensive solution. Organizations must integrate various AI applications, such as summarization, chatbots, and data analytics, to create effective workflows. This piecemeal approach requires careful orchestration and expertise.

Ethical and Regulatory Considerations

The webinar briefly touches on concerns about whether AI-generated proposals are acceptable within public procurement rules. While no widespread bans exist, the lack of clear policies creates uncertainty. Organizations must navigate evolving regulations and societal expectations regarding AI use.

Future Trends and Ethical Considerations in AI for Public Procurement

AI is expected to further transform public sector sales and procurement by enabling smarter automation, deeper insights, and more strategic, data-driven decision-making.

AI-Generated RFPs and Bid Evaluations

AI not only assists in responding to requests for proposals (RFPs) but also creates them. This could build a circular ecosystem where AI tools generate RFPs, bids, and evaluations. While intriguing, this raises questions about fairness, transparency, and the human role in procurement processes.

Price-Only Evaluations and Market Impacts

There is a growing trend toward price-only tender evaluations for standard goods and services. While efficient, this approach risks undermining quality and innovation, particularly for complex or creative projects. AI might exacerbate this by standardizing proposals, making differentiation more challenging.

Human-AI Collaboration and the Role of Creativity

Human skills like human creativity, strategic thinking, and domain expertise remain irreplaceable. AI should be seen as an enabler that automates routine tasks and enhances decision-making, but does not supplant the unique contributions of skilled professionals.

Industry Adoption and Societal Acceptance

Public sector organizations vary in their readiness to embrace AI. Early adopters like the Publications Office of the European Union are integrating AI to improve service efficiency. Wider acceptance will depend on demonstrating trust, transparency, and tangible benefits.

Frequently Asked Questions

Is AI currently allowed in public sector proposal writing?

There is no general prohibition against using AI in proposal writing for public tenders. However, organizations should be mindful of transparency and ensure that AI-generated content is reviewed and validated by humans.

Can AI analyze all tender documents, including confidential ones?

Most AI tools currently focus on publicly published tenders. Confidential documents, such as second-stage procedures or internal requests, require different handling and are often not accessible for AI analysis.

How accurate are AI-generated tender summaries?

AI tender summarization tools like those used by Hermix achieve high accuracy rates (around 98-99%) for extracting key information. Nonetheless, final decisions should always involve human verification.

Can AI help with pricing strategies in tenders?

AI can provide price trend analysis based on available data, but it is limited by the confidentiality of financial proposals. Pricing decisions should consider market intelligence alongside human judgment.

What is prompt engineering, and why is it important?

Prompt engineering involves crafting precise and effective questions or commands to AI systems to get accurate and useful responses. It is an emerging skill critical to maximizing AI’s value in public sector sales.

Are there risks of AI replacing jobs in the public sector sales?

AI automates routine tasks but is unlikely to replace roles that require creativity, strategic thinking, and human judgment. Instead, it enables professionals to focus on higher-value activities.

What if everyone starts using AI? Won’t all proposals look the same?

That’s a valid concern. Generic tools may produce similar outputs. The difference lies in how tools are used and how human experts personalize and strategize on top of AI-generated content.

What about AI hallucinations? Can we trust the outputs?

Hermix has performed extensive testing and validation. In our experience, hallucinations are nearly non-existent, with accuracy rates of 95–98%. Still, we always recommend a human-in-the-loop approach.

How do you differentiate from general tools like ChatGPT?

Hermix combines cutting-edge AI with domain specific expertise in public procurement. While general purpose tools lack business context, Hermix is built specifically for sales and bid professionals.

Conclusion

The integration of artificial intelligence into public sector sales and procurement represents a profound shift in how companies approach tendering and contract management. As demonstrated by the insightful discussion between Stefan Morcov and Rudolf de Schipper, AI-powered tools are already enabling faster, smarter, and more strategic decision-making. From big data analytics and graph-based market intelligence to GPT-driven document summarization and bid assistance, the capabilities are advancing fast.

While AI is powerful, it is not a silver bullet. Winning public tenders still requires strategic planning, creativity, and deep understanding of both the client and the market. AI can generate a good proposal, but not a winning one,  that remains a human achievement.

At Hermix, we take a pragmatic approach: combining advanced AI and LLM tools with deep public sector sales experience. Our goal is not to replace professionals, but to help them work faster, better, and smarter and ultimately, to help companies win more public contracts.

For those interested in exploring these technologies further, Hermix offers free trial accounts and personalized demonstrations, inviting companies to experience firsthand the power of AI in public procurement analytics and automation.

This article was created following the AI for public sector sales automation and analytics live webinar, hosted by Rudolf de Schipper and Stefan Morcov.