Sourcing Intelligence: Why 89% of B2B buyers rely on AI – and still seek human expertise

Man versus machine? The perfect setup for the global procurement market

Global B2B procurement is undergoing an unprecedented transformation. A toxic mix of geopolitical tensions, vulnerable supply chains, and stringent ESG requirements is forcing companies to radically rethink their sourcing strategies. Artificial intelligence (AI) is positioning itself as the supposed savior in this volatile era, promising rapid data analysis, enormous cost savings, and fully automated processes in seconds. The prevailing narrative is that those who ignore this technological leap will be left behind. However, the euphoria surrounding algorithmic omnipotence reveals a dangerous blind spot. AI systems smooth out nuances, filter out essential empirical data, and fail precisely where it matters most in the complex world of procurement: in building genuine trust and assessing unforeseen crises. This article explores why machine perfection can quickly become a competitive disadvantage, why true authenticity is the currency of the future, and how the strategic synthesis of data-driven AI and human judgment forms the foundation for successful global sourcing of tomorrow.

Why human expertise remains irreplaceable in the global B2B procurement market – and why polished AI perfection becomes a competitive disadvantage

The new area of ​​tension: data machines versus market intelligence

Global B2B procurement has undergone more change in the past three years than in the previous two decades. The convergence of pandemic-related supply chain disruptions, the rapid maturation of generative AI, stricter ESG regulations, and a fundamental generational shift in procurement departments has unleashed a dynamic that leaves no company untouched. Digital platforms promise fully automated supplier matching in hours instead of weeks, AI systems analyze millions of data points in real time, and autonomous purchasing agents negotiate offers without human intervention. Those who continue to rely on purely analog processes in this environment are undeniably losing ground.

But this euphoria surrounding algorithmic omnipotence creates a blind spot that can prove costly for companies in global sourcing. AI systems smooth out differences, level personalities, and produce a frictionless average consensus. Those who rely solely on machine-generated procurement intelligence risk losing precisely what matters in volatile markets: the ability to make context-based judgments, cultivate relationships, and interpret signals that no data set captures.

The topography of the global procurement market in 2026

The structural forces shaping the global procurement market today are multifaceted and, in some cases, contradictory. On the one hand, there is China's continued dominance: despite tariff threats and geopolitical tensions, two-thirds of companies worldwide plan to maintain or even expand their business with China in 2025. China plays a key role, particularly in rare earth elements and raw materials for digitalization and the energy transition; for refinery products, Germany and the EU currently have little recourse to China. This is not a short-term dependency, but a structural foundation that, despite European counter-movements, can only be shifted slowly.

On the other hand, commodity markets are under continued pressure. Geopolitical tensions, structural shifts, and high costs continue to shape global commodity markets. The copper market experienced extreme price fluctuations in the second quarter of 2025: After falling to $8,540 per ton in April, the price reached a yearly high of $10,100 per ton in June – a spike that directly reflects the trade escalation caused by US tariffs of up to 50 percent on copper imports. Aluminum is operating in a similar volatile environment: Global inventories in June 2025 were around 67 percent lower than the previous year, while geopolitical developments and US tariffs are causing additional market distortions.

This volatility is not a temporary phenomenon. For commodity procurement, it means that price and currency risks are increasing in parallel, and decisions must be made under greater time pressure. Under these conditions, real-time information and data analysis tools are becoming increasingly important for making informed and flexible decisions. However, real-time data is not self-explanatory; it requires interpretation.

Nearshoring, Friendshoring and the new geography of trust

When asked how companies are dealing with this fragility, a clear answer is emerging: through the geographical restructuring of their supply chains. In light of geopolitical crises, 80 percent of consumer goods and retail companies in Germany are again focusing more on regional sourcing, and 83 percent are investing in so-called friendshoring – concentrating on suppliers in politically allied countries. In practice, nearshoring often means relocating production capacities to Eastern Europe, Turkey, or North Africa, resulting in significantly shorter delivery times and increased responsiveness, but also placing new demands on border processes, customs clearance, and infrastructure.

This friendshoring is far more than a logistical adjustment. It's a geopolitical risk decision that deeply impacts core business operations. Reorganizing supply chains along political axes of trust requires a foundation of regional knowledge, networks, and cultural competence that no algorithm can spontaneously provide. Diversifying suppliers to reduce dependence on individual regions and countries is a strategic response to destabilized global supply chains—and it presupposes knowing whom to trust. Trust isn't built on data points, but on experience.

European policymakers are responding with the Critical Raw Materials Act: With minimum quotas of 10 percent for domestic extraction of strategic raw materials, 40 percent for processing, and 25 percent for recycling by 2030, the EU is setting clear benchmarks for self-sufficient raw material supply. Large companies with more than 500 employees and over €150 million in annual revenue have been required since May 24, 2025, to conduct a risk assessment of their raw material supply chain every three years. This creates structural compliance requirements that demand in-depth analysis and market knowledge – not mere data aggregation.

What AI can actually achieve in the procurement process

The power of AI in procurement is real and impressive. Next-generation AI systems use large language models to understand procurement requirements contextually, employ graph databases to map supplier relationships, and continuously improve matching quality through reinforcement learning from user feedback. What once took weeks—from requirements definition and supplier identification to shortlisting—can now be accomplished in hours. 74 percent of procurement managers plan to increase their automation investments by 2026, and automation can reduce cycle times by up to 50 percent.

In the area of ​​cost optimization, AI delivers tangible results. According to a BCG analysis, consistent use of AI can achieve savings of up to 5 percent in direct procurement and 15 percent in indirect procurement. AI reduces procurement costs by identifying inefficient spending, supporting dynamic pricing, and strengthening negotiations with suppliers. Through real-time monitoring and predictive analytics, AI detects potential supplier risks early, enabling proactive disruption management. B2B companies benefit from up to 50 percent higher closing rates through AI-supported implementation—provided the quality of the underlying data is sufficient. This last caveat is crucial.

AI automates time-consuming tasks such as research, analysis, contract review, and invoice reconciliation. It improves decision quality through pattern recognition in large procurement datasets, supports more accurate forecasts, and facilitates early risk assessments. Procurement teams can better evaluate supplier relationships because AI continuously monitors performance, reliability, and risks. The added value is evident and should not be underestimated.

The systematic limitations of machine-based procurement intelligence

Despite these performance indicators, AI in B2B procurement encounters structural limitations that are often underestimated in practice. The first and most fundamental limitation concerns the ability to form judgments in situations lacking historical precedents. AI can analyze, structure, summarize, and formulate information—but true orientation only arises through conscious thought and human judgment. In negotiations where reputation, relationship history, and cultural contexts play a role, algorithms merely represent the average behavior of past transactions.

The second limitation is the phenomenon of algorithmic leveling. Generative AI systems strive for neutrality, smoothing out differences until only a superficial average remains. In procurement platforms that use AI for supplier recommendations, this leads to strong differentiating characteristics being systematically filtered out. For the algorithm, anything without a structured data point simply doesn't exist. Companies relying on AI-generated recommendation lists thus regularly miss out on suppliers who, while lacking a perfect digital presence, possess rare market knowledge or privileged supplier networks.

The third boundary concerns trust and relationship building. Seventy percent of B2B buyers prefer suppliers with clear, open communication, especially in times of uncertainty. This kind of trust isn't built on technology alone, but on transparent processes and responsible data handling. In B2B procurement decisions, which often involve significant investments and long-term commitments, 72 percent of decision-makers consult at least three different reference sources before shortlisting a new supplier. This vetting process is an inherently human one: talking to colleagues, consulting experts, and evaluating personal experiences.

Finally, there is a fourth, less discussed limitation: data quality dependency. If the quality of the input data is poor, even the most sophisticated AI will produce flawed recommendations. Around 18 percent of B2B providers still see no concrete applications for AI within their organizations. While the democratization of advanced procurement intelligence through AI creates new opportunities—especially for small and medium-sized enterprises—it also presents challenges in the areas of data quality, costs, skills gaps, and ethical considerations that must be carefully addressed.

State-of-the-art cargo planes, optimized transport routes, and multimodal logistics chains are interchangeable—they can be bought, leased, or outsourced. What money can't buy are direct contacts with producers in Peruvian mines, reliable supply relationships in the CIS countries, and years of built-up trust in markets that are unfamiliar to outsiders. The decisive competitive advantage in global commodity trading lies not in transporting the good from A to B, but in knowing where the good comes from, who produces it, and how to gain access before others even know the market exists. Whoever owns the network sets the price. Everyone else pays it.

More information here:

• Integrated Sourcing & Trading House: Raw Materials, Global Sourcing & Trading

Authenticity as a competitive factor: What polished AI language destroys

While AI undoubtedly achieves efficiency gains in operational procurement, a new problem arises in the area of ​​market positioning and trust building: the inflation of interchangeable content. The more companies use AI-generated texts, supplier evaluations, and communication modules, the more homogenous the information environment becomes—and the more valuable those who speak with their own judgment, genuine experience, and a clear personality become.

According to a recent study by Nosto, 86 percent of consumers say that authenticity is crucial when deciding which brands to support. This dynamic is even more pronounced in the B2B context. B2B purchasing decisions are complex, long-term, and involve high costs. Trust determines whether orders are awarded, risk tolerance is assumed, and recommendations are made. Authenticity, reliability, and expertise are essential for sustainable customer relationships. In a world where even market leaders can become invisible due to AI systems because their product data is hidden in PDFs or contradictory information exists between their website and press releases, consistent substance becomes a strategic advantage.

Positioning claims that aren't backed up by corresponding achievements and genuine expertise are quickly exposed as unbelievable. The reverse is also true: those who actually possess rare expertise in specific industry fields and communicate it with genuine style, rather than hiding it behind AI-smoothed language, achieve a differentiation that cannot be replicated algorithmically. Honesty and transparency are essential for building trust, and customers quickly discern whether someone is truly committed to a collaborative partnership or simply using optimized language.

The strategic configuration of Sourcing Intelligence: Human and Machine

The real question isn't whether AI or human expertise is better in global sourcing. It's how both elements can be configured to complement each other optimally. 71 percent of companies plan to collaborate more closely with IT sourcing consultants in the future, partly to better represent their interests to cloud providers. This reflects the fundamental understanding that digital transformation doesn't function optimally without human leadership and oversight.

The most productive approach looks like this: AI takes over the data-intensive, repetitive, and speed-critical tasks—market price monitoring, supplier databases, early risk warnings, and compliance checks. Human expertise handles context-dependent interpretation, relationship building, strategic classification, and final judgment. Responsibility remains human, because every decision has consequences, and consequences always affect people. This division of labor is not a temporary state on the path to complete automation—it is a permanent model for complex markets.

While B2B buyers use generative AI tools as a starting point for research, they are increasingly turning to peers, experts, and the suppliers themselves to validate the results of these tools. This shift is fundamental: AI can handle the initial level of information gathering, but decisions in challenging sourcing situations—negotiating during commodity shortages, switching suppliers in politically sensitive regions, assessing long-term reliability—require what AI cannot structurally provide: embedded knowledge derived from lived experience in specific industries and markets.

Industry expertise as a non-replicable advantage

What is often overlooked in the current debate about AI disruption is that domain-specific know-how in niche industrial markets cannot be replaced by training data. Mechanical engineering, energy infrastructure, intralogistics – these are fields where market developments, regulatory signals, and technological trajectories require years of analysis before any assessments are reliable. Raw material markets for critical minerals such as lithium, cobalt, or rare earths follow geopolitical logics that become obsolete faster than any historical data set.

B2B procurement in these sectors is built on trust. Lengthy decision-making processes involving multiple decision-makers on the client side demand in-depth analysis. Inconsistencies between different communication channels can quickly undermine the credibility of the positioning. Consistency – in language, judgment, and attitude – cannot be generated algorithmically; it is the result of genuine conviction and sound expertise. In the energy sector, for example, the decision is not made by the supplier with the best SEO profile, but by the one whose expertise is trusted and who is believed to act appropriately even in unforeseen situations.

Added to this is the team dimension. A well-coordinated team of specialists from various B2B domains – mechanical engineering, energy, digital, logistics – can establish connections that remain invisible to a single specialist or a purely data-driven system. Cross-functional expertise is the raw material for sourcing intelligence in the truest sense: not mere data processing, but networked thinking across industry, technology, and market boundaries.

Visibility in the era of algorithmic preselection

Another aspect that is increasingly putting companies in the B2B market under pressure: 89 percent of B2B buyers already use AI in their procurement process. For them, anyone missing from the results simply doesn't exist. A recent study by TrustRadius shows that 72 percent of decision-makers encounter AI-powered overviews during their research, and 90 percent of them access the cited sources to verify the information. This means: The first selection stage is algorithmic, the second is human – and it is precisely in this second stage that genuine substantive content is decisive.

Generative AI systems aim to remain neutral, smoothing out discrepancies to achieve a factual average. For procurement experts and platforms with genuine depth in niche markets, this presents an opportunity, not a threat. Those who possess structured, substantive, and precise content on specific topics—commodity markets, trading platform comparisons, mechanical engineering suppliers, ESG compliance—will be cited preferentially by AI systems, outperforming generalists with superficial content. Visibility in the AI ​​era is not a matter of budget, but a matter of depth.

ESG, compliance and the new dimension of ethical procurement

Regulatory developments have fundamentally changed the requirements for global procurement. The EU Critical Raw Materials Act, the CSDDD, and the Uyghur Forced Labor Prevention Act in the US—these regulations obligate companies to actively monitor and ensure transparency in their supply chains, far beyond traditional supplier audits. Digitized supply chains are twice as transparent and 30 percent more punctual than their non-digitized counterparts, but budget constraints and shifting priorities are hindering the progress of many companies.

The hidden danger lurks not in the known, but in the unseen: ongoing sanctions between the EU and China, sudden supply chain disruptions, dependencies on raw materials that can become unavailable during political tensions, and rising cyber risks in critical infrastructure. A Chief Procurement Officer tasked with anticipating these invisible risks, modeling scenarios, and establishing proactive procurement strategies needs more than just a dashboard. Silence is not a sign of security, but a warning signal. Here, too, human judgment is irreplaceable—not because AI cannot generate scenarios, but because weighing the consequences of actions is an act of responsibility that cannot be delegated.

Sustainability in the supply chain is seen as a competitive advantage by 83 percent of German companies – however, only 57 percent have launched corresponding initiatives to actually implement this aspiration. This gap between aspiration and reality is characteristic of a transformation phase in which operational requirements still overshadow strategic commitments.

The synthesis: Sourcing intelligence as a combination of data and judgment

What practice teaches us is both sobering and inspiring: Neither side – neither the purely data-driven machine nor the isolated expert – can deliver the quality that the global procurement market, in its current complexity, demands. Synthesis is the only viable path. AI provides speed, data depth, and scalability. Human expertise provides context, trust, and the ability to correctly interpret the unexpected.

Sourcing intelligence, in its truest sense, is therefore not a technology, but a competency – an organizational capability that combines structured data analysis with sound market understanding, genuine networks, and clear values. This combination cannot be arbitrarily reproduced. It develops over time, through experience in specific markets, through mistakes and corrections, through established relationships, and through in-depth industry knowledge. In an era where AI systems are capable of automating generic procurement services in minutes, the lasting competitive advantage lies not in automation itself, but in what cannot be automated: authentic competence, personality, and the well-established interplay of diverse domain expertise within a team.

Companies that understand this use AI for what it is: a powerful tool in experienced hands. Nothing more, but also nothing less.

I would be happy to serve as your personal advisor.

Konrad Wolfenstein

Email: wolfenstein@xpert.Digital

LinkedIn

 

 

Industry focus areas: B2B, digitalization (from AI to XR), mechanical engineering, logistics, renewable energies and industry

More information here:

• Expert Business Hub

A thematic hub offering insights and expertise:

• Knowledge platform covering global and regional economies, innovation and industry-specific trends
• A collection of analyses, insights, and background information from our key areas of focus
• A place for expertise and information on current developments in business and technology
• A hub for companies seeking information on markets, digitalization, and industry innovations