# Digital Transformation Concept Card: [Programme, company or course case] > A concept card for the technology strategy elective: what digital transformation is and is not, why technology > pays off only with complementary changes in processes, skills and organisation (Brynjolfsson and Hitt 2000; > Brynjolfsson, Rock and Syverson 2021), the levels of IT-enabled change from automation to business scope > redefinition (Venkatraman 1994), digital maturity (Westerman, Bonnet and McAfee 2014), the building blocks of a > digital company (Ross, Beath and Mocker 2019), dynamic capabilities for renewal (Teece, Pisano and Shuen 1997; > Warner and Wäger 2019), and the finance of a transformation: adoption curves, the J-curve of cash flows, stage-gated > funding and a business-model shift from selling products to selling a service. It builds on four core cards and > links to them instead of re-teaching them: `mba/mba-ai-strategy-card` (where AI creates value), > `mba/mba-change-management-card` (Kotter, ADKAR, resistance), `mba/mba-value-chain-card` (activities and fit) and > `mba/mba-npv-irr-card` (NPV, IRR, payback). It is for second-year MBA students and for managers who sponsor or > govern a transformation programme. Source basis: the original works cited below, written in original words. Every > company and number is fictional (Northfield Outdoor Co., Tallow Creek Medical Center, Corvane Robotics) and was > computed in Python. This is a learning aid, not legal, investment or regulatory advice. Replace every [bracketed] > value and every row marked *Example*. ## Document control | Field | Value | |---|---| | Course, case or programme | [Technology strategy elective, session 8 / unified commerce programme] | | Learner, analyst or programme office | [Your name or team] | | Executive sponsor | [Name and role] | | Business case version and date | [v0.3, YYYY-MM-DD] | | Discount rate source | [mba/mba-capm-wacc-card / company hurdle rate memo, date] | | Currency and unit | [USD m] | | Spreadsheet used | [Excel / Google Sheets / LibreOffice Calc] | | Reviewed by | [Instructor, mentor, finance business partner or architecture board] | | Last updated | YYYY-MM-DD | ## Contents - [When to use it](#when-to-use-it) - [The idea in plain language](#the-idea-in-plain-language) - [Frameworks in more depth](#frameworks-in-more-depth) - [Formulas and notation](#formulas-and-notation) - [Assumptions and limits](#assumptions-and-limits) - [How managers use it in practice](#how-managers-use-it-in-practice) - [Worked example A (fully solved): Northfield unified commerce programme](#worked-example-a-fully-solved-northfield-unified-commerce-programme) - [Worked example B (half solved): Tallow Creek online scheduling, big bang or pilot](#worked-example-b-half-solved-tallow-creek-online-scheduling-big-bang-or-pilot) - [Exercise C (unsolved): Corvane Robotics from selling robots to robots as a service](#exercise-c-unsolved-corvane-robotics-from-selling-robots-to-robots-as-a-service) - [Connections to other cards](#connections-to-other-cards) - [Common mistakes and misconceptions](#common-mistakes-and-misconceptions) - [Self-test questions](#self-test-questions) - [My notes and next steps](#my-notes-and-next-steps) - [Sources and further reading](#sources-and-further-reading) - [Answer key](#answer-key) - [Revision history](#revision-history) - [Standard and licence note](#standard-and-licence-note) ## When to use it **One-sentence rule:** use this card when a firm plans a multi-year, cross-functional change in how it creates and captures value with digital technology, and needs to decide what to change, in what order, how to fund it and how to know whether it is working. | Use it when | Use something else when | |---|---| | A board approves a "digital" or "unified commerce" programme spanning several functions | You are scoring individual AI use cases (`mba/mba-ai-strategy-card`) | | A case asks why a large IT investment did not raise productivity | You need the people side of one change in depth (`mba/mba-change-management-card`) | | Management must choose between a big-bang rollout and a staged pilot | You need a decision tree with several chance nodes (`mba/mba-decision-trees-card`) | | A product company considers selling outcomes or subscriptions instead of products | You need platform and network-effect economics (`mba/mba-platform-strategy-card`) | | An architecture or programme board must sequence initiatives and legacy retirement | You need the full portfolio of innovation bets (`mba/mba-innovation-portfolio-card`) | ## The idea in plain language **What it is.** Vial (2019), reviewing a large body of research, defines digital transformation as a process that aims to improve an entity by triggering significant changes to its properties through combinations of information, computing, communication and connectivity technologies. In his framework, digital technologies disrupt the environment, the organisation responds strategically, and the response changes how it creates value, which in turn requires structural changes and meets organisational barriers. Two parts of this view matter in practice. First, the change is in the **business**: how customers are served, how work is done, how money is made. A new system that leaves processes and roles unchanged is a technology upgrade, not a transformation. Second, it is a **process**, not a project with an end date: firms that "finish" a transformation and return to business as usual tend to fall behind again. **Digitisation, digitalisation, transformation.** Practitioners use three words that are easy to confuse. Digitisation converts analogue information into digital form (scanning paper forms). Digitalisation uses digital technology to change a process (online appointment booking replaces the phone queue). Transformation changes the firm's value proposition or business model (a robot maker selling throughput per hour instead of robots). Each level needs more organisational change than the one before. **Complements decide the return.** Brynjolfsson and Hitt (2000) summarised evidence that the payoff from information technology depends on complementary investments: redesigned processes, decentralised decision rights, new skills, new incentives. The same software installed in two firms can produce very different results because one firm changed how it works and the other did not. Milgrom and Roberts (1990) gave the economic logic: when practices are complements, adopting one raises the return to the others, so partial adoption can be worse than either the old system or the full new one. This is the strongest single reason transformation business cases fail: they budget the technology and assume the complements. **The J-curve.** Brynjolfsson, Rock and Syverson (2021) argue that general purpose technologies (Bresnahan and Trajtenberg 1995) require large intangible investments (new processes, data, training, organisational redesign) that standard accounts expense or do not measure. Measured productivity therefore dips first and rises later: a J-curve. Inside one firm the same shape shows up in cash flows. Spending comes first, adoption is slow, benefits arrive in years three to five. A steering committee that judges the programme on year-one benefits will cut it at the bottom of the J. **From automation to new scope.** Venkatraman (1994) described five levels of IT-enabled business transformation: localised exploitation (one function automates), internal integration (systems and processes connected across functions), business process redesign, business network redesign (processes rebuilt with suppliers, partners and customers) and business scope redefinition (the firm changes what business it is in). The first two are evolutionary and the last three revolutionary: potential benefits rise with each level, and so does the degree of change required. **Adoption is a behaviour, not a go-live date.** A system "goes live" on one day; people adopt it over months. The technology acceptance model (Davis 1989) explains individual adoption by perceived usefulness and perceived ease of use. A programme that measures go-live dates and not adoption will report success while benefits stay on paper. The change side of adoption (sponsorship, resistance, reinforcement) is covered in `mba/mba-change-management-card`. ## Frameworks in more depth **Strategic alignment.** Henderson and Venkatraman (1993) proposed that IT value depends on fit across four domains: business strategy, IT strategy, organisational infrastructure and processes, and IT infrastructure and processes. They showed that alignment can be driven from different starting points, for example strategy driving technology, or a new technology opening a new strategy. Bharadwaj, El Sawy, Pavlou and Venkatraman (2013) went further and argued that in a digital business the IT strategy and the business strategy fuse into one **digital business strategy**, with four themes: its scope (what it covers, including partner ecosystems), its scale (how fast it can grow through networks and cloud capacity), its speed (of launches, decisions and network formation) and the sources of value creation and capture (data, multi-sided business models, control of architecture). **Digital maturity.** Westerman, Bonnet and McAfee (2014), based on a survey and interviews with large firms, sorted companies on two dimensions: **digital capability** (what the firm does with technology in customer experience, operations and business models) and **leadership capability** (how it leads: vision, employee engagement, governance and the relationship between IT and the business). The four types are Beginners (low on both), Fashionistas (many digital initiatives, weak leadership and coordination), Conservatives (strong governance, cautious and slow on digital capability) and Digital Masters (high on both). They reported that Digital Masters were more profitable than their peers. Use the grid for diagnosis and for the conversation it starts, not as a causal claim about your firm. | Type | Digital capability | Leadership capability | Typical risk | First move | |---|---|---|---|---| | Beginners | Low | Low | Unaware of the threat | Build a shared vision and a few visible wins | | Fashionistas | High | Low | Duplicated, uncoordinated initiatives and spend | Governance: one roadmap, shared platforms, kill duplicates | | Conservatives | Low | High | Governance blocks experiments | Fund fast pilots inside the guard rails | | Digital Masters | High | High | Complacency | Keep renewing; protect the next horizon | **Building blocks of a digital company.** Ross, Beath and Mocker (2019) separate an **operational backbone** (reliable, integrated core systems and processes for the existing business) from a **digital platform** (reusable data, business and technology components that teams can combine quickly into new offerings). Around them sit shared customer insights (what customers will pay for, learned by experiment), an accountability framework (who owns which component and outcome) and an external developer platform (components opened to partners). Their practical lesson: firms with a messy backbone cannot build digital offerings quickly, because every new offering needs custom integration with the old systems. Legacy cleanup is often the first, least glamorous phase of a transformation. **Dynamic capabilities.** Teece, Pisano and Shuen (1997) defined dynamic capabilities as the firm's ability to integrate, build and reconfigure competences in a changing environment. Warner and Wäger (2019), from interviews with incumbent firms, described digital transformation as an ongoing process of strategic renewal built on digital sensing (scouting technologies and customer change), digital seizing (rapid prototyping, balancing a portfolio of bets) and digital transforming (redesigning internal structures, improving digital maturity). The connection to this card: a firm that only runs one transformation programme has a project; a firm that can sense, seize and transform repeatedly has a capability. **Technical debt and the legacy estate.** Cunningham (1992) introduced the debt metaphor: shortcuts in software are like borrowing, and the "interest" is the extra effort every later change costs until the debt is repaid. In a transformation case the interest appears as higher run costs, slower releases and integration effort for every new initiative. Retiring legacy systems is a source of savings (Worked example A includes it) and a precondition for speed. Track it in `architecture/tech-debt-register` and rationalise the application estate with `architecture/application-portfolio-rationalization`. **Regulation as a design constraint (dated fact).** Digital programmes that collect product, customer or patient data must design for the rules that apply where they operate. One example for products sold in the EU: the Data Act (Regulation (EU) 2023/2854) entered into force on 11 January 2024 and has applied since 12 September 2025, according to the European Commission's Data Act page (checked October 2026, https://digital-strategy.ec.europa.eu/en/policies/data-act). It gives users of connected products rights to access and share the data those products generate; the design obligation in Article 3(1) (data accessible to the user by default) applies only to connected products and related services placed on the market after 12 September 2026 (Article 50(3)), so products already on the market are not redesigned retroactively. *Note on sources:* the Article 50 wording here was checked on published reproductions of the Official Journal text, because EUR-Lex could not be opened at checking time. Check the authentic text on EUR-Lex (https://eur-lex.europa.eu/eli/reg/2023/2854/oj) and take advice from counsel before you rely on any of this. For AI components, see `mba/mba-ai-risk-governance-card`. ## Formulas and notation **Adoption curve** (logistic S-curve, a convenient planning shape, not a law): $$ a_t = \frac{1}{1 + e^{-k\,(t - t_0)}}, \qquad 0 < a_t < 1 $$ **Benefit and cash flow in year $t$:** $$ B_t = a_t \, s \, B^{\text{full}}, \qquad CF_t = B_t + L_t - R_t - I_t $$ **Value, payback and the J-curve:** $$ NPV = \sum_{t=0}^{T} \frac{CF_t}{(1+r)^t}, \qquad \text{cumulative } C_\tau = \sum_{t=0}^{\tau} CF_t, \qquad \text{payback} = \min\{\tau : C_\tau \ge 0\} $$ **Stage-gated funding** (pilot first, roll out only on success, rollout one year later): $$ E[NPV_{\text{big bang}}] = -C + p\,PV_s + (1-p)\,PV_f, \qquad E[NPV_{\text{pilot}}] = -C_p + p\,\frac{\max(0,\;PV_s - C)}{1+r} $$ **Maturity index** (weighted score on each axis): $$ M = \sum_{j} w_j \, x_j, \qquad \sum_j w_j = 1, \qquad x_j \in [1, 5] $$ **Complementarity** (Milgrom and Roberts 1990): technology $T$ and organisational practice $O$ are complements when the gain from one is larger when the other is present, $$ V(T{=}1, O{=}1) - V(T{=}0, O{=}1) \;>\; V(T{=}1, O{=}0) - V(T{=}0, O{=}0). $$ | Symbol | Meaning | Unit | |---|---|---| | $a_t$ | Share of the target user base using the new way of working in year $t$ | 0 to 1 | | $k$ | Steepness of the adoption curve (higher = faster adoption) | per year | | $t_0$ | Year in which adoption reaches 50% | year | | $B^{\text{full}}$ | Annual benefit at full adoption (margin uplift plus cost savings) | USD m | | $s$ | Share of the full benefit the design can deliver (below 1 when complements are missing) | 0 to 1 | | $L_t$ | Legacy run cost avoided after decommissioning | USD m | | $R_t$ | Run cost of the new platform (licences, cloud, product teams) | USD m | | $I_t$ | Investment: technology, data, process redesign, training | USD m | | $r$ | Discount rate (`mba/mba-capm-wacc-card`) | % per year | | $C$, $C_p$ | Cost of full rollout, cost of a pilot | USD m | | $p$ | Probability that the solution works as designed at scale | 0 to 1 | | $PV_s$, $PV_f$ | Present value of benefits if the solution succeeds, if it fails | USD m | | $w_j$, $x_j$ | Weight and score of maturity dimension $j$ | weight, 1 to 5 | | $V(T, O)$ | Value with ($=1$) or without ($=0$) technology $T$ and practice $O$ | USD m | ## Assumptions and limits - **The S-curve is a planning shape.** Real adoption stalls, jumps after a policy change, or splits by segment (stores adopt faster than warehouses). Model segments separately when they differ. - **Benefits are estimates with optimism bias.** Use base rates from the firm's own past programmes and haircut revenue benefits more than cost savings. Assign each benefit an owner who will carry it in a budget (`project/benefits-realization-plan`). - **Intangible investment is real investment.** Training, process redesign and data cleanup are often expensed and cut first. The complementarity logic says cutting them can destroy more value than it saves. - **Stage-gating assumes the pilot is informative.** If a pilot cannot tell success from failure (the wrong site, too short, no control group), its option value is small. Design pilots as experiments (`product/ab-test-plan`). - **Maturity scores are self-assessments.** Calibrate them with evidence (adoption data, release frequency, share of revenue from digital channels) and with people outside the programme office. - **NPV misses strategic options.** A digital platform may enable offerings no one has yet imagined. Value obvious follow-on options with `mba/mba-options-real-options-card`; do not use "strategic value" to rescue a weak case. - **Regulation, security and privacy are constraints, not afterthoughts.** Data rules, cybersecurity and AI rules change; record the date you checked each and who confirmed it. ## How managers use it in practice 1. **Frame the ambition** in business terms: which customer, process or business-model change, which value driver (`mba/mba-value-chain-card`) and which Venkatraman level. 2. **Diagnose** maturity, backbone health and technical debt. Decide what must be fixed before new offerings can scale. 3. **Design the complements** next to the technology: process changes, decision rights, roles, skills, incentives. Budget them explicitly. 4. **Build the business case** with an adoption curve, run costs, legacy savings and a J-curve view of cumulative cash, then a sensitivity on adoption speed. 5. **Choose the funding mode**: big bang or stage-gated, with clear evidence required at each gate. 6. **Run the change**: sponsor, coalition, communication and reinforcement (`mba/mba-change-management-card`). 7. **Measure adoption and benefits**, not only delivery milestones, and feed results back into the next gate. ```mermaid flowchart TD A["Frame the ambition: customer, process or business model"] --> B["Diagnose maturity, backbone and technical debt"] B --> C{"Is the operational backbone fit for the change?"} C -- No --> D["Phase 0: fix backbone and retire legacy"] C -- Yes --> E["Design technology and complements together"] D --> E E --> F["Business case with adoption curve and J-curve"] F --> G{"Is success uncertain and can a pilot reveal it cheaply?"} G -- Yes --> H["Stage-gated: pilot, evidence gate, rollout"] G -- No --> I["Committed rollout with adoption milestones"] H --> J["Change plan: sponsor, coalition, reinforcement"] I --> J J --> K["Track adoption, benefits, run cost each quarter"] K --> L{"Benefits on track?"} L -- Yes --> M["Scale and fund the next horizon"] L -- No --> N["Fix complements or stop and redeploy"] N --> E ``` ## Worked example A (fully solved): Northfield unified commerce programme **Situation (fictional).** Northfield Outdoor Co. runs stores and an e-commerce site on separate systems: stores cannot see online stock, the website cannot sell store stock, and markdowns are set by region without demand data. The executive team proposes a **unified commerce programme**: a single inventory view, ship-from-store and click-and-collect, demand-based markdowns and one customer profile. This is internal integration and business process redesign in Venkatraman's terms. Discount rate 9%, horizon seven years, no terminal value (a deliberately conservative choice for a programme whose platform will need renewal). **Step 1: diagnose maturity.** Northfield's leadership team scored itself with evidence from a two-week review (scores 1 to 5): | Axis | Dimension | Weight | Score | Weighted | |---|---|---|---|---| | Digital capability | Customer experience *(Example)* | 0.40 | 2.5 | 1.000 | | Digital capability | Operations *(Example)* | 0.35 | 2.0 | 0.700 | | Digital capability | Business model *(Example)* | 0.25 | 1.5 | 0.375 | | | **Digital capability index** | 1.00 | | **2.075** | | Leadership capability | Vision *(Example)* | 0.30 | 4.0 | 1.200 | | Leadership capability | Engagement *(Example)* | 0.25 | 3.5 | 0.875 | | Leadership capability | Governance *(Example)* | 0.25 | 3.0 | 0.750 | | Leadership capability | IT-business relationship *(Example)* | 0.20 | 3.0 | 0.600 | | | **Leadership capability index** | 1.00 | | **3.425** | Excel: `=SUMPRODUCT(C2:C4,D2:D4)` for each axis. With a threshold of 3.0 on both axes, Northfield is a **Conservative**: good governance and a clear vision, weak digital capability. The first move is to fund fast, well-governed change in operations, not to add governance. **Step 2: cash flows of the full programme.** Investment USD 6.0m, 5.0m and 2.0m in years 0 to 2, of which USD 1.0m a year is process redesign and training (the complements). Full-adoption benefit $B^{\text{full}}$ = USD 8.0m a year (fewer markdowns 2.5, ship-from-store margin 3.0, store process savings 1.5, fewer split shipments 1.0). Run cost USD 1.5m a year from year 1. Legacy systems retired at the end of year 2 save USD 0.8m a year from year 3. Adoption follows the S-curve with $k = 1.2$ and $t_0 = 3$, $s = 1$. | Year | Adoption $a_t$ | Benefit | Legacy saving | Run cost | Investment | Cash flow | Cumulative | |---|---|---|---|---|---|---|---| | 0 | 0.0000 | 0.00 | 0.00 | 0.00 | 6.00 | -6.00 | -6.00 | | 1 | 0.0832 | 0.67 | 0.00 | 1.50 | 5.00 | -5.83 | -11.83 | | 2 | 0.2315 | 1.85 | 0.00 | 1.50 | 2.00 | -1.65 | -13.48 | | 3 | 0.5000 | 4.00 | 0.80 | 1.50 | 0.00 | 3.30 | -10.18 | | 4 | 0.7685 | 6.15 | 0.80 | 1.50 | 0.00 | 5.45 | -4.73 | | 5 | 0.9168 | 7.33 | 0.80 | 1.50 | 0.00 | 6.63 | 1.90 | | 6 | 0.9734 | 7.79 | 0.80 | 1.50 | 0.00 | 7.09 | 8.99 | | 7 | 0.9918 | 7.93 | 0.80 | 1.50 | 0.00 | 7.23 | 16.22 | Year 0 has no benefit: the programme is being built, so adoption is set to 0 in year 0 instead of the curve value (0.0266). Excel, with year 0 in row 2 and the year in column A: adoption from year 1 `=1/(1+EXP(-1.2*(A3-3)))`, benefit `=B3*8`, cash flow `=C3+D3-E3-F3`, cumulative `=H2+G3`. | Measure | Value | Excel | |---|---|---| | NPV at 9% | USD 6.16m | `=NPV(9%, G3:G9)+G2` | | IRR | 19.1% | `=IRR(G2:G9)` | | Payback (undiscounted) | 4.71 years | `=4+4.73/6.63` | | Discounted payback | 5.48 years | year 5 cumulative discounted cash flow -2.02, year 6 discounted cash flow 4.23: `=5+2.02/4.23` | | Lowest point of the J-curve | USD -13.48m at the end of year 2 | `=MIN(H2:H9)` | **Step 3: the technology-only case.** The finance team asks why the programme cannot save the USD 3.0m of process redesign and training. Without the complements, assume adoption is slower ($k = 0.6$, $t_0 = 5$) and the design delivers only 60% of the full benefit ($s = 0.6$), because stores keep old routines and the markdown team overrides the demand model. Investment falls to USD 5.0m, 4.0m and 1.0m. | Year | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | |---|---|---|---|---|---|---|---|---| | Adoption $a_t$ | 0.0000 | 0.0832 | 0.1419 | 0.2315 | 0.3543 | 0.5000 | 0.6457 | 0.7685 | | Cash flow | -5.00 | -5.10 | -1.82 | 0.41 | 1.00 | 1.70 | 2.40 | 2.99 | | Cumulative | -5.00 | -10.10 | -11.92 | -11.51 | -10.51 | -8.81 | -6.41 | -3.42 | NPV = USD -6.01m (`=NPV(9%, C3:I3)+B3`), IRR -6.5%, no payback within seven years. Saving USD 3.0m of complements (present value USD 2.76m, `=1+1/1.09+1/1.09^2`) loses USD 12.18m of NPV (6.16 minus -6.01). This is the complementarity argument in numbers: the "cheaper" case is the more expensive one. ```vega-lite { "$schema": "https://vega.github.io/schema/vega-lite/v5.json", "description": "Northfield (fictional): cumulative cash flow of the unified commerce programme, USD m", "width": 380, "height": 220, "data": {"values": [ {"year": 0, "case": "Full programme", "cumulative": -6.0}, {"year": 1, "case": "Full programme", "cumulative": -11.83}, {"year": 2, "case": "Full programme", "cumulative": -13.48}, {"year": 3, "case": "Full programme", "cumulative": -10.18}, {"year": 4, "case": "Full programme", "cumulative": -4.73}, {"year": 5, "case": "Full programme", "cumulative": 1.9}, {"year": 6, "case": "Full programme", "cumulative": 8.99}, {"year": 7, "case": "Full programme", "cumulative": 16.22}, {"year": 0, "case": "Technology only", "cumulative": -5.0}, {"year": 1, "case": "Technology only", "cumulative": -10.1}, {"year": 2, "case": "Technology only", "cumulative": -11.92}, {"year": 3, "case": "Technology only", "cumulative": -11.51}, {"year": 4, "case": "Technology only", "cumulative": -10.51}, {"year": 5, "case": "Technology only", "cumulative": -8.81}, {"year": 6, "case": "Technology only", "cumulative": -6.41}, {"year": 7, "case": "Technology only", "cumulative": -3.42} ]}, "layer": [ { "mark": {"type": "line", "point": true}, "encoding": { "x": {"field": "year", "type": "quantitative", "title": "Year", "axis": {"tickMinStep": 1}}, "y": {"field": "cumulative", "type": "quantitative", "title": "Cumulative cash flow (USD m)"}, "color": {"field": "case", "type": "nominal", "title": null} } }, { "data": {"values": [{"zero": 0}]}, "mark": {"type": "rule", "strokeDash": [4, 4]}, "encoding": {"y": {"field": "zero", "type": "quantitative"}} } ] } ``` **Step 4: recommendation.** (1) Approve the full programme, complements included, and ring-fence the USD 3.0m of process redesign and training so it cannot be cut to "save" money. (2) Tell the board the J-curve plainly: cumulative cash bottoms at USD -13.5m at the end of year 2 and turns positive during year 5; the year-1 review should judge adoption, not benefits. (3) Track adoption by store each quarter against the curve; if adoption at the end of year 2 is below 15% (against 23% planned), stop and fix the complements before spending more. (4) Retire the legacy order and stock systems on schedule; the USD 0.8m a year saving is part of the case and slips easily. ## Worked example B (half solved): Tallow Creek online scheduling, big bang or pilot **Situation (fictional).** Tallow Creek Medical Center plans online self-scheduling for outpatient clinics, replacing a phone-and-fax process. Discount rate 7%. A full rollout to all clinics costs USD 4.0m. If the design works (patients use it, clinic templates hold, no-shows do not rise) the present value of benefits is USD 10.0m; if it does not, staff fall back to phones and the present value of benefits is only USD 1.5m. The team puts the probability of success at 0.55. Alternative: a one-year pilot in three clinics for USD 0.5m that reveals whether the design works; roll out a year later only if it does. | Item | Value | Excel | |---|---|---| | NPV of big bang if success | 6.00 | `=10-4` | | NPV of big bang if failure | [ ] | `=1.5-4` | | $E[NPV_{\text{big bang}}]$ | [ ] | `=-4+0.55*10+0.45*1.5` | | NPV of rollout after a successful pilot, valued today | 5.607 | `=(10-4)/1.07` | | $E[NPV_{\text{pilot}}]$ | [ ] | `=-0.5+0.55*5.607` | | Value of the pilot (difference) | [ ] | `=E_pilot-E_bigbang` | | Highest pilot cost at which the pilot still wins | [ ] | `=0.5+difference` | 1. Which route has the higher expected NPV? [ ] 2. Above what probability of success does the big bang win? Set the two expressions equal and solve for $p$. [ ] 3. Name two things the expected NPVs leave out (patient experience during the pilot, staff fatigue from two changes, learning that improves the design). [ ] 4. What must the pilot measure for it to be informative? Name at least three metrics. [ ] ## Exercise C (unsolved): Corvane Robotics from selling robots to robots as a service Corvane Robotics sells autonomous mobile robots for USD 120k each. Each robot costs USD 78k to build. Sixty per cent of buyers also take a service contract worth USD 10k a year at a 50% margin, for five years. Corvane considers a **robots-as-a-service** offer instead: the customer pays USD 3.6k a month (paid annually in arrears for this exercise) on a three-year contract; Corvane keeps ownership. Corvane's cost to serve is USD 5k a year of service and USD 2k a year of software and monitoring. The team expects 80% of customers to renew for years 4 and 5; ignore any residual value of the robot. Discount rate 12%. 1. Compute the NPV per robot of the sale with service contract. Give the Excel formula. 2. Compute the cash flows and the NPV per robot of robots as a service, and its IRR. 3. At what monthly fee does robots as a service match the NPV of the sale? 4. Corvane places 500 robots in year 0. What is the cash effect at year 0 under each model? What does that mean for funding (link `mba/mba-venture-finance-cap-table-card` and `mba/mba-capital-structure-card`)? 5. Which of Venkatraman's five levels is this move? Which complements (sales incentives, service operations, monitoring data, finance) must change for it to work? 6. Give two strategic reasons Corvane might prefer the service model even if its NPV per robot is lower. ## Connections to other cards - `mba/mba-ai-strategy-card`: AI use cases are often the most visible part of a transformation; score them there. - `mba/mba-change-management-card`: the adoption curve in this card is the result of the change plan in that card. - `mba/mba-value-chain-card`: locate each digital initiative in an activity and check fit across activities. - `mba/mba-npv-irr-card` and `mba/mba-options-real-options-card`: the valuation tools behind Worked example A and B. - `mba/mba-decision-trees-card`: extend the pilot decision to several gates and imperfect pilot signals. - `mba/mba-platform-strategy-card`: when the transformation creates a platform with network effects. - `mba/mba-innovation-portfolio-card` and `mba/mba-growth-strategy-card`: balance core, adjacent and transformational bets. - `mba/mba-customer-lifetime-value-card`: the subscription economics behind Exercise C. - `mba/mba-technology-strategy-guide`: the elective course this card belongs to. - `architecture/business-capability-map` and `architecture/current-state-assessment`: the diagnosis step in practice. ## Common mistakes and misconceptions 1. **Treating transformation as an IT project.** If no process, role or business model changes, it is an upgrade. 2. **Budgeting the technology and assuming the complements.** Training, process redesign and data cleanup decide the return (Worked example A). 3. **Switching benefits on at go-live.** Adoption takes years; model it with a curve and track it. 4. **Judging the programme at the bottom of the J.** Year-one benefits are small by design; judge leading indicators. 5. **Building new offerings on a broken backbone.** Every new offering then needs custom integration, and speed never arrives. 6. **Running many uncoordinated digital initiatives.** The Fashionista pattern: activity without shared platforms or governance. 7. **Using "strategic value" to rescue a negative NPV.** Name the specific options and value them, or drop the claim. 8. **Pilots that cannot fail.** A pilot chosen at the best site with no control group gives no information and no option value. 9. **Ignoring the cash and risk shift of a service model.** Selling outcomes moves capital and performance risk onto the seller's balance sheet (Exercise C). 10. **Counting legacy savings before the legacy is retired.** Run both until switch-off; the saving starts only then. 11. **Treating regulation as a final check.** Data, privacy and AI rules shape the design; check them early and date your check. ## Self-test questions 1. Distinguish digitisation, digitalisation and digital transformation with one example each. 2. What did Brynjolfsson and Hitt (2000) conclude about the payoff from IT, and why does it matter for a budget? 3. Explain the productivity J-curve (Brynjolfsson, Rock and Syverson 2021) in two sentences. 4. Name Venkatraman's (1994) five levels and say which are evolutionary and which revolutionary. 5. Name the four types in Westerman, Bonnet and McAfee's (2014) maturity grid and the first move for a Fashionista. 6. What is the difference between an operational backbone and a digital platform (Ross, Beath and Mocker 2019)? 7. With $k = 1.2$ and $t_0 = 3$, what is adoption in year 2? Give the Excel formula. 8. A programme has cumulative cash flow of -4.73 at the end of year 4 and a year-5 cash flow of 6.63. What is the payback period? 9. When is a stage-gated pilot worth more than a big-bang rollout? Name two conditions. 10. Define complementarity in Milgrom and Roberts' (1990) sense with an example from a store. 11. Name the three clusters of dynamic capabilities for digital transformation in Warner and Wäger (2019). 12. Why does a move from product sales to a service model change a firm's funding needs? ## My notes and next steps | Date | Note or action | Done | |---|---|---| | YYYY-MM-DD | *Example:* Score my employer on the maturity grid with two colleagues and compare our scores | [ ] | | YYYY-MM-DD | *Example:* Rebuild Worked example A with $k = 0.8$ and find the new payback year | [ ] | | YYYY-MM-DD | [Your action] | [ ] | ## Sources and further reading - Bharadwaj, A., El Sawy, O. A., Pavlou, P. A., and Venkatraman, N. (2013). Digital business strategy: Toward a next generation of insights. *MIS Quarterly*, 37(2), 471-482. - Bresnahan, T. F., and Trajtenberg, M. (1995). General purpose technologies: "Engines of growth"? *Journal of Econometrics*, 65(1), 83-108. - Brynjolfsson, E., and Hitt, L. M. (2000). Beyond computation: Information technology, organizational transformation and business performance. *Journal of Economic Perspectives*, 14(4), 23-48. - Brynjolfsson, E., Rock, D., and Syverson, C. (2021). The productivity J-curve: How intangibles complement general purpose technologies. *American Economic Journal: Macroeconomics*, 13(1), 333-372. - Cunningham, W. (1992). The WyCash portfolio management system. Experience report, OOPSLA 1992. - Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. *MIS Quarterly*, 13(3), 319-340. - Henderson, J. C., and Venkatraman, N. (1993). Strategic alignment: Leveraging information technology for transforming organizations. *IBM Systems Journal*, 32(1). - Milgrom, P., and Roberts, J. (1990). The economics of modern manufacturing: Technology, strategy, and organization. *American Economic Review*, 80(3). - Ross, J. W., Beath, C. M., and Mocker, M. (2019). *Designed for Digital: How to Architect Your Business for Sustained Success*. MIT Press. - Teece, D. J., Pisano, G., and Shuen, A. (1997). Dynamic capabilities and strategic management. *Strategic Management Journal*, 18(7), 509-533. - Venkatraman, N. (1994). IT-enabled business transformation: From automation to business scope redefinition. *Sloan Management Review*, 35(2). - Vial, G. (2019). Understanding digital transformation: A review and a research agenda. *Journal of Strategic Information Systems*, 28(2), 118-144. - Warner, K. S. R., and Wäger, M. (2019). Building dynamic capabilities for digital transformation: An ongoing process of strategic renewal. *Long Range Planning*, 52(3), 326-349. - Westerman, G., Bonnet, D., and McAfee, A. (2014). *Leading Digital: Turning Technology into Business Transformation*. Harvard Business Review Press. - Regulation (EU) 2023/2854 (Data Act), European Commission policy page, https://digital-strategy.ec.europa.eu/en/policies/data-act (checked October 2026). - Textbooks (further reading, by name only): Laudon and Laudon, *Management Information Systems*; Pearlson, Saunders and Galletta, *Managing and Using Information Systems: A Strategic Approach*; Rogers, *The Digital Transformation Playbook*. - Open resource: OpenStax, *Principles of Management* (CC BY-NC-SA 4.0), chapters on organisational change and innovation, https://openstax.org/details/books/principles-management. ## Answer key **Worked example B.** NPV of big bang if failure = 1.5 - 4.0 = USD -2.50m. $E[NPV_{\text{big bang}}]$ = -4.0 + 0.55 x 10.0 + 0.45 x 1.5 = USD 2.175m. $E[NPV_{\text{pilot}}]$ = -0.5 + 0.55 x 5.607 = USD 2.584m. Value of the pilot = 2.584 - 2.175 = USD 0.409m. Highest pilot cost at which the pilot still wins = 0.5 + 0.409 = USD 0.909m. (1) The pilot. (2) Setting $-4 + 10p + 1.5(1-p) = -0.5 + 5.607p$ gives $8.5p - 2.5 = 5.607p - 0.5$, so $p = 2.0 / 2.893 = 0.691$: above a 69.1% probability of success the big bang wins, because the cost of waiting a year then outweighs the protection the pilot gives (check: at $p = 0.70$ big bang 3.450, pilot 3.425; at $p = 0.85$ big bang 4.725, pilot 4.266). (3) Examples: patient and staff experience during a dual-process year, the risk that a pilot at three clinics does not represent all clinics, the learning that improves the design before rollout (which would raise $PV_s$), and staff fatigue from two waves of change. (4) Examples: share of appointments booked online, no-show rate against a matched control clinic, call volume, staff time per booking, patient satisfaction, booking errors and accessibility complaints. **Exercise C.** (1) Sale: year 0 margin 120 - 78 = 42, then 0.6 x 10 x 0.5 = USD 3.0k a year for five years. NPV = 42 + 3.0 x 3.6048 = USD 52.81k per robot (`=42+PV(12%,5,-3)`). (2) Robots as a service: fee 3.6 x 12 = USD 43.2k a year, net of cost to serve 43.2 - 7.0 = 36.2. Cash flows: -78 in year 0, 36.2 in years 1 to 3, 0.8 x 36.2 = 28.96 in years 4 and 5. NPV = USD 43.78k (`=NPV(12%,36.2,36.2,36.2,28.96,28.96)-78`), IRR 33.9% (`=IRR(range)`). Undiscounted total 88.52 against 57.0 for the sale: the service earns more in total but later. (3) Break-even fee USD 3.824k a month (Goal Seek on the NPV difference). (4) Sale: +USD 21.0m at year 0 (500 x 42k). Service: -USD 39.0m (500 x 78k) before any fee is received. The model turns a cash-generating sale into a capital-intensive rental book that needs equity, debt or asset-backed funding. (5) Business scope redefinition, or at least business network redesign: Corvane changes what it sells. Complements: sales commissions on contract value, not units; service operations with uptime targets; fleet monitoring data; finance for asset funding and revenue recognition; pricing for renewals. (6) Examples: recurring revenue and closer customer relationships (data on robot use improves products and service), lower purchase barrier for customers without capital budgets, higher renewal-based lifetime value if renewal exceeds 80%, a barrier to switching once fleets and software are integrated. **Self-test.** 1. Digitisation: scanning paper delivery notes. Digitalisation: online booking replacing phone booking. Transformation: a robot maker selling throughput as a service. 2. The return to IT depends on complementary organisational investments (process redesign, decision rights, skills); a budget that funds only technology underfunds the source of value. 3. General purpose technologies need large unmeasured intangible investment first, so measured productivity falls before it rises. The same logic gives a firm-level cash flow J-curve. 4. Localised exploitation and internal integration (evolutionary); business process redesign, business network redesign and business scope redefinition (revolutionary). 5. Beginners, Fashionistas, Conservatives, Digital Masters. Fashionistas need governance: one roadmap, shared platforms, fewer duplicated initiatives. 6. The backbone runs the existing business reliably on integrated core systems; the digital platform is a set of reusable components for building new offerings fast. 7. $a_2 = 1/(1 + e^{1.2}) = 0.2315$; `=1/(1+EXP(-1.2*(2-3)))`. 8. $4 + 4.73/6.63 = 4.71$ years. 9. When success is uncertain, a pilot can reveal it cheaply and reliably, and the cost of delay is small relative to the loss avoided on failure (Worked example B: below $p = 0.691$). 10. Two practices are complements when each raises the return to the other. Example: a store stock-visibility system and a new rule that store staff pick online orders; the system is worth little without the rule, and the rule is unworkable without the system. 11. Digital sensing, digital seizing, digital transforming. 12. The seller keeps the asset and is paid over time, so it funds the asset up front (Exercise C: -USD 39.0m for 500 robots instead of +USD 21.0m) and carries performance and renewal risk. ## Revision history | Version | Date | Author | Change summary | |---|---|---|---| | 0.1 | YYYY-MM-DD | [Your name] | First draft from the template | | 0.2 | YYYY-MM-DD | [Your name] | *Example:* Adoption at month 18 below plan; $k$ lowered from 1.2 to 0.9 and payback re-estimated | ## Standard and licence note Based on the published research on information technology, organisation and strategy (Henderson and Venkatraman 1993; Venkatraman 1994; Teece, Pisano and Shuen 1997; Brynjolfsson and Hitt 2000; Bharadwaj et al. 2013; Westerman, Bonnet and McAfee 2014; Vial 2019; Ross, Beath and Mocker 2019; Warner and Wäger 2019; Brynjolfsson, Rock and Syverson 2021) and on the European Commission's public Data Act page as checked in October 2026. The frameworks are described in own words; the text, examples and numbers are original and fictional. The card has no affiliation with any school, publisher, consulting firm or author named. Check the original works and the current official texts before you rely on them. Template licence: MIT.