The Knowledge No AI Model Knows: How Organizations Learn from Each Other

INTERORGANIZATIONAL LEARNING · TACIT KNOWLEDGE · L&D STRATEGY

When everyone in your industry uses the same AI tools: what still sets your organization apart?

The question is uncomfortable because the answer doesn’t lie in the technology. Same models, same vendors, similar prices. What remains is the knowledge that is documented nowhere: why a project failed three years ago, how a rollout really moves through the organization, what resistance appears when theory meets the works council. This experiential knowledge, called “tacit knowledge” in research, is in no training dataset. It emerges inside organizations, and it travels in only one way: through people who pass it on.

This is exactly where interorganizational learning comes in. And this is exactly where it is becoming relevant again.

Closing off is the wrong response to uncertain times

The signals from global politics are clear: trade wars, protectionism, national go-it-alone strategies. Following that logic, organizations should guard their knowledge like treasure. Research and practice on knowledge sharing tell a different story: isolation hinders innovation and prevents progress.

While AI reshapes entire industries, geopolitical tensions tear supply chains apart and economic pressure forces rethinking, three questions arise anew. How do we develop solutions to problems we don’t yet fully understand? How do we stay able to act when the rules change daily? How do we build resilience without losing touch?

What sets interorganizational learning apart from networking

Interorganizational learning (IOL) happens whenever people from different organizations share experiences with one another. This is how Nataša Rupčić described it in 2021 in the journal “The Learning Organization”. Participants take in new ideas, information and best practices from others, build their own competence from them and change their organization.

The difference from networking is not one of degree.

IOL therefore requires a high degree of transparency and willingness to adapt. Participants must be ready not only to exchange information but to rethink their own ways of working and to build new competencies together. That is more demanding than a networking evening, and it is more effective.

What the research shows

In 2018, Anni Rajala of the University of Vaasa statistically combined more than 20 studies on the relationship between IOL and performance. The result across all studies: organizations that use IOL improve their innovation performance, meaning their ability to develop new products, services and processes. And they improve their market performance, such as market share and sales.

A second finding is even more revealing for practice. In 2015, Brian Connelly and colleagues examined more than 37,000 interorganizational relationships and distinguished whether partners trust each other because of competence or because of integrity. Collaboration is ten times more efficient when organizations trust each other because of their integrity. Competence remains important, but for collaboration, integrity is far more decisive.

On top of that, IOL produces results that individual organizations could not have achieved alone or would have had to buy on the market as consulting services.

Flavio Garcea, Head of AI Delivery Hub at Clearstream (Deutsche Börse Group), who came into contact with more than 70 other organizations through ada on the topic “Tech and Transformation”, describes how this works in practice: “The most valuable moments in the ada Fellowship were seeing how people from completely different industries and sectors approached tasks differently. It often surprised me in a positive way and showed me that there are other approaches.”

Organizations benefit twice over: from successful approaches and from the mistakes of others. Both considerably reduce their own implementation risk.

The learning dilemma: how much do we give away?

Every organization that wants to learn across organizational boundaries faces the same tension. How much knowledge do I reveal without losing competitive advantage? How open am I to outside knowledge without losing my own identity? A Swedish research group around Rikard Larsson described this question more than 25 years ago as the “learning dilemma”.

Their answer is unspectacular, and therefore useful: in a world where all organizations are maximally transparent and receptive, the benefit would be maximized for everyone. Because nobody can ever be completely certain of the other party’s intentions, a compromise is the best solution: a good balance of moderate transparency and moderate receptiveness.

A second risk remains: newly acquired knowledge sometimes can’t be applied in one’s own context. In a 2019 meta-analysis, Rosileia Milagres and Ana Burcharth showed that knowledge is harder to transfer the more implicit, complex, technological, collective and systemic it is. Anyone planning IOL should therefore think about transfer into their own organization from the very beginning.

The most common worry, by contrast, mostly doesn’t come true in practice: that employees see the greener grass of the neighboring organization and leave. Rebekka Scholz, Team Lead Industry & Field Marketing at T-Systems International, describes the effect as more or less the opposite: “The exchange in the programs strengthened my positive impression of T-Systems. I realized how advanced my company is in many technological areas. This outside view strengthened my commitment to the company and the industry.”

Four conditions under which IOL works

1. Strategically anchored rather than well-intentioned: Before choosing a format comes the question of the goal. Should innovation be driven forward, or is the aim to strengthen the employer brand and develop potential? In the first case, the format needs room for substantive discussion and prototypes; in the second, more visibility, networking and influence. Without this decision, there is no yardstick for success later.
2. Measure impact, not satisfaction: “Of course, employee satisfaction and recommendation are metrics for the success of an interorganizational program. But what matters far more is which impact can be measured in the organization, for example the implementation rate of innovation projects or the identification of new talent who can transform the company from within,” says Alexander Lorbeer, former CEO of ada Learning GmbH.
3. Motivation beats hierarchy: People who are intrinsically eager for change also bring what they have learned back into the company. What matters is the ability not only to take in new knowledge but to apply it in practice and integrate it into existing processes. In research this is called “absorptive capacity”, empirically confirmed in a 2008 meta-analysis by Raymond van Wijk and colleagues. Just as important is the internal networking of participants. Without multipliers in their own organization, even the best external learning remains ineffective.
4. The right altitude: When very different levels of competence and experience meet, the result is frustration instead of learning. “Conversations go very differently depending on whether it’s about long-term talent development strategy or the operational rollout of a learning software,” says Johanna Voigt, Principal Consultant for Learning & Development at HRpepper. Strategic and operational questions therefore need separate rounds.

Effective interorganizational learning is therefore not a stroke of luck but the result of systematic design: the right people, well-prepared structures, reliable trust.

This article is based on the German-language specialist article “Wie Organisationen voneinander lernen” (How Organizations Learn from Each Other) by Dr. Laura E. Creon and Hannah Wolters, published on September 26, 2025 in the trade magazine “Neues Lernen” (Haufe).