Transparency for Value Creation, Fairness, and Quality of Life

In the self-driving company, software makes the majority of operational and tactical decisions. That only works if every one of these decisions can be traced in detail. This article shows why transparency therefore begins with the software, how it aligns the organization with the overall optimum, how it creates fairness toward employees, partners, and the state, and why it ultimately brings more quality of life.

Transparency begins with the software

In my book "Das selbstfahrende Unternehmen" (Springer Gabler 2021, in German) I described how the classic end-to-end processes turn into networked, multidimensional functions that communicate with each other continuously and in real time; in The End of Processes, Long Live the Algorithms I explored that idea further. One consequence of this networking is often overlooked: a company whose functions are fully represented in software becomes, for the first time in its history, fully visible. Every decision has a trigger, a data basis, and a rule by which it was made, and all of that can be stored, queried, and audited.

This traceability is not a pleasant side effect; it is a precondition. When algorithms decide, management must be able to intervene when unexpected changes occur, and to do so it must understand why the system decided the way it did. People adopt a defensive attitude toward algorithms they do not understand; whatever is not understood gets pushed into the realm of the mystical. Acceptance only emerges with traceability. In How Self-Driving Companies Decide I showed that the vast majority of decisions in companies are rule-based and can therefore be explained exactly. Where learning methods are used, this applies all the more: the decision logic must be specified, the data basis documented, the result logged.

It is no coincidence that this topic is in the spotlight right now. Since ChatGPT became available to everyone last November, companies have been debating which decisions they want to entrust to software and how they verify its results. And yesterday, on June 14, 2023, the European Parliament adopted its negotiating position on the AI Act; negotiations with the Council are now beginning. What the final regulation will look like remains open. For the self-driving company, the direction is clear regardless of the outcome: traceability is not a concession to the legislator but the foundation that allows software to take on responsibility in the first place.

An end to departmental thinking

Anyone who has worked for a longer period in a larger, hierarchically structured organization knows the pattern: over time, every department develops its own logic and optimizes primarily for itself in order to look better than the other departments. Information is withheld, mistakes are papered over, incomplete or incorrect data is passed along. The damage often goes unnoticed for years and only surfaces once those responsible have long since left the company.

Complete transparency ends this game because it makes the key figures of every function visible in real time and immediately attributes every deviation to the overall system. The benefit for the organization as a whole thus becomes the only maxim that still holds, and the company can align its decisions with the overall optimum to a degree that departmental boundaries never allow. The struggles for promotion that are fought in classic companies through flattery and intrigue also lose their basis; the energy that was tied up there flows into productive work.

This explicitly does not mean that people become mere executors of the system. On the contrary: the routines that used to consume their time disappear, and they can turn to the tasks that genuinely require judgment. As described in The Organizational Structure for True Agility, these people work in largely autonomous teams that the system supplies with the most current information, aligned with the overall optimum. In the spirit of complete transparency, all results can be inspected at any time with regard to every decision criterion.

Fairness toward employees, partners, and the state

Transparency works in three directions at once: inward toward employees, outward toward suppliers and customers, and toward the state as the tax and supervisory authority.

Careers based on results, not on performance theater

Because middle, operational, and tactical management is replaced by software, careers become more fact-oriented and flatter. What counts is whether a team actually achieves the agreed key results in line with the overarching goals, not how convincingly someone puts on a show. Thanks to the high level of transparency, compensation can be aligned with performance, and roles come with more personal responsibility. Even unpleasant decisions, such as dismissals, are justified by the system transparently, fairly, and traceably. For those affected, this is a considerable difference from the opaque procedures that are common today.

Trading platforms that treat small partners fairly, too

Externally, transparency emerges through shared trading and contract platforms that organize sales for one side and procurement for the other. Their added value lies in the automated exchange of order, delivery, and invoice data: instead of transferring every partial amount back and forth individually, the system continuously keeps the balance of mutual credits and receivables. In the book I played this through using the fictional company GRANOBIZ in the year 2035: whereas large corporations in the 2020s extracted advantages for themselves with tax tricks, the absolute transparency of such a platform provides a fairness from which even small suppliers such as regional organic farmers benefit. Suppliers that deliver unreliably or cannot keep up technologically, on the other hand, quickly lose their place in the market: every service disruption forces the self-driving company back into the analog world at that point, where humans once again have to fix things by hand.

Taxes in real time

Toward the state, transparency goes furthest. Since all data is continuously available, the self-driving company can pay its taxes and levies in real time. The relevant laws then no longer exist merely as text but as algorithms that are made available to the company as a computational core. Because this core works reliably and transparently, the tax administration can rely on all levies being paid correctly and on time. Routine audits then no longer need auditors, because the system itself is fully verifiable; tax advisors concentrate on structural questions and act as management consultants. Companies will continue to try to reduce their tax burden with ingenious arrangements. The difference: under complete transparency, only arrangements that are one hundred percent correct and withstand any audit remain an option.

Game theory explains why this is more than control: those who know that the other parties also stick to the rules are themselves considerably more willing to act correctly. Transparency thus creates a fairness that rests not on distrust but on certainty.

Humanitarian and ecological principles become verifiable

The road to the self-driving company can be traveled with confidence because every step can be assessed with regard to its consequences; the data and facts needed for that are available in real time, with a transparency that did not exist before. And because the system makes the operational and tactical decisions, people gain the time and the headspace to attend to humanitarian and ecological concerns and to improve them.

A self-driving company must not only be run humanely, it must also act in a humanitarian way; in the book I formulated this as a condition for the company of 2035. The exploitation of people and the overexploitation of nature are already under criticism today and will be tolerated even less in the future. The reason once again lies in transparency: where all data is open to inspection, questionable practices can be concealed neither internally nor externally. Embellished sustainability reports and greenwashing lose their effect because the data on which they rest is out in the open. In the end, only companies that actually adhere to ecological and humanitarian principles retain access to the markets.

Above all, the absolute transparency of self-driving companies toward customers, employees, the state, and society will increasingly ensure that all forces are bundled in the right direction.

From "Das selbstfahrende Unternehmen" (Springer Gabler 2021), own translation

Quality of life for executives and owners

The extreme transparency of the company relieves those at the top of many pseudo-decisions. Liquidity, capacity utilization, order situation, the structure of markets and customer groups, forecasts, and simulations on a reliable data basis: all of it is available at any time, without anyone having to request a report. The work of executives and owners concentrates on far-reaching strategic decisions based on excellent data; planning horizons shift to more than ten years. The roles of ownership and management move closer together because, thanks to transparency and continuous forecasts, the company is easier to steer.

When the majority of operational and tactical decisions is automated, nobody needs 80-hour weeks with ten meetings a day to keep the company under control. The self-driving company thus also provides for the quality of life of its executives and owners. Those who want to work more and earn more can still do so; nobody takes that freedom away. And for customers, the combination of falling unit costs and individualized products means a greater variety of better and yet affordable products.

What companies can do now

The road to this transparency does not begin with a law or with a new technology, but with decisions within your own organization. From our projects we know five starting points that can be implemented regardless of industry and size:

1. Make the decision logic explicit. Which rules underlie a credit approval, a price change, or a reorder? As long as these rules exist only in the heads of clerks, no software can apply them traceably. A specification that describes rules, data, and exceptions is the first step.
2. One data basis instead of many truths. Transparency fails when sales, production, and finance work with different figures. The data of every function must be readable by all other functions in real time.
3. Log decisions. Every automated decision needs an entry: trigger, data state, applied rule, result. Only this audit trail makes management interventions and third-party audits possible.
4. Explainability before sophistication. Where decisions affect people, a traceable set of rules is preferable to an opaque model, even if the latter would be more accurate in individual cases. Learning methods belong where their results can be verified and explained; in The Diversity of Intelligent Software Systems I described which methods are suited to which tasks.
5. Replace documents with data. Purchase order, order confirmation, delivery note, invoice, and payment confirmation are carriers of data that has long existed electronically. Whoever recognizes demand automatically, records the service, and triggers payment with the electronic invoice has closed the process chain without paper and without retyping, and in a way that keeps every step transparent and traceable in detail.

None of these steps requires the self-driving company. Each of them, however, already delivers a noticeable share of its effect today: less coordination effort, faster decisions, and a data basis that management, partners, and auditors can all rely on equally.

The next step

Transparency is the precondition for making autonomy, sustainability, resilience, and humanity, the four factors of the self-driving company, verifiable in the first place. To see how the vision is structured as a whole, visit the book page: About the book. If you would like to know where your company stands in terms of the traceability of its decisions and which functions can be mapped transparently in software first, we would be happy to talk with you: Book an expert meeting.