The Sixty Effects Nobody Wrote Down

Most digital transformation proposals are argued on one or two expected effects. Faster processing. Lower cost. Better customer data. The proposal is approved on that basis, measured on that basis, and declared a success or a failure on that basis. Researchers at the University of Potsdam approached the question from the opposite end. Rather than asking what companies _intended_, they documented what actually happened across 75 real companies of varying size from 40 industries — financial services, retail, telecommunications, manufacturing, agriculture — that had already undergone digital transformation. They identified 60 distinct impacts, which they then aggregated inductively into 10 impact types distributed across three dimensions: the value creation model, the value proposition model, and the customer interaction model. Two observations from that work deserve a moment of executive attention before any of the detail. The first is that impacts arrive on their own schedule: some are intended and some are not, some become apparent immediately and others surface only years later. The second is that a single digital intervention produced not one effect but a cluster of them, which could correlate with or cross-fertilise one another. The implication is uncomfortable but useful. A transformation project is not an investment with a return; it is an intervention with consequences. The business case describes the intended fraction. The remainder arrives regardless of whether anyone anticipated it, and the research suggests the remainder is usually larger.

The value is in the specificity, so it is worth reading the map rather than summarising it. Under the value creation model sit four impact types. _Process alignment_ covers transparency for management, speed, efficacy, efficiency, integration of formerly separate processes, automation that replaces human labour, meeting legal requirements, shifts in how far customers participate in the process, and the tying of upstream supplier processes into internal ones. _Staff_ covers addressing skill shortages, employee empowerment in decision-making, creativity — which the authors note may be fostered _or contained_ — inspiration, the changing importance of staff safety and wellbeing, and new approaches to recruiting. _Exploitation of data_ covers new KPIs and testing made possible by digitisation, digital twins mirroring physical objects and their states, information gain for employees and customers alike, customer behaviour data, market data, real-time data, remote monitoring of objects, facilities or persons, and ubiquitous data accessible without constraints of time or place. _Networks_ covers interaction, cooperation, platform-based trade between stakeholders, and network effects. Under the value proposition model sit two. _Business development_ covers effects on the offline core business, business model adjustment toward new data-based models, growth or decline in prospects, realignment of organisational structure, outsourcing decisions, and three distinct forms of dependency — on resources, on third-party providers, and on structural circumstances. _Product development_ covers product innovation and creation, integration and aggregation into bundles or platforms, design, customisation to individual needs, cost reduction for the customer, product quality, continuous update functionality, interoperability, add-ons enabling up- and cross-selling, and acting as a digital enabler for one's own customers. Under the customer interaction model sit four: _customer behaviour_ (attitude toward the brand, and how purchasing decisions are made), _customer relations_ (pre-sales and after-sales service quality, support, user experience, community building, retention, and lock-in within an ecosystem), _channel management_ (the style and the mode of communication, interaction and transaction), and _marketing_ (the marketing mix and the brand itself). Beneath all ten sit five underlying causes: ubiquitous sensors and actors, automated communication and networking, data-processing capabilities, advanced machine-learning techniques, and enhanced human-computer interaction — themselves drawn from a technology categorisation of 22 characteristics across communication technologies, intelligent hardware-software systems, and data technologies.

The authors are explicit about how practitioners should use this: as a reference and as a creativity tool — to define a deliberate technology mix, to develop the future business model systematically along the three dimensions, to assess the company's current transformation status, and to see the spectrum of implications a future initiative might carry. Read as a mirror, the ten types make an unusually honest review instrument. Take a completed initiative and walk it through all ten; the types it touched without appearing in any planning document are the organisation's blind spots, and they tend to repeat. Read as a map, the same list becomes a design tool, and the dependency and staff entries repay particular attention. The framework is deliberately neutral about direction — the core business may be impaired _or_ promoted, creativity fostered _or_ contained, dependency increased _or_ mitigated — which is a more candid posture than most consulting frameworks adopt, and a more useful one for governance. A caveat belongs here: this is a working paper, and the authors present the model as something that gives _direction_ to exploring cause-effect relationships rather than establishing them. It maps the possibility space; it does not predict which cells your organisation will land in. That is the work, and it is worth beginning with a simple question put to your last completed project: of these ten, how many did it touch — and how many appeared anywhere in the document that authorised it?