Transforming a Corporate R&D Centre
R11 is an insider account of one central laboratory rebuilding itself around business-unit strategy, with performance figures attached and no method section. The model is worth having. The causal claim it carries is not proven, and the paper's own numbers show you where the joins are.
What R11 is, and what a single case can carry
R11 is a 2006 practitioner-journal account of how a large electronics manufacturer restructured its corporate R&D centre between 2000 and the mid-2000s. One author had previously worked in the studied organisation's chief technology office; the other was a serving executive of the parent group at the time of writing. There is no description of data collection, sampling, interview protocol or analysis anywhere in the paper.
That combination — insider authorship, performance statistics, no method — is best read as a practitioner account with numbers attached rather than as independent research. It does not make the model useless. It changes what the model can be used to argue.
Four periods of drift
The paper's first framework is a chronology. It matters because it names the failure state the transformation was designed to escape, and because the third period is the counterintuitive one.
- 1987 to early 1990sFounded to prepare future businesses and support existing ones with advanced technologies. It could not maintain a unique status among the group's other R&D centres, which mostly ran short-term applications work. In uncomfortable relationships with business units, it spent heavily on short-term projects to satisfy them — 80 percent of its R&D projects were allocated to applied development — which created conflict with the business-unit centres.
- Early to mid 1990sThe group adopted a new management philosophy, redirecting cost-oriented management to quality-oriented management. From 1993 this penetrated all business lines and gave the centre an opening to repair the relationship. With inefficient management systems and relatively low technological capability, it still struggled with its primary mission.
- Mid to late 1990sStructure changed to research laboratories organised around major research fields, focusing on long-term and basic projects rather than short-term development, with more attention to recruiting talent. The management system became more efficient but remained inefficient at increasing technological capability and creating new business opportunities.
- 2000 onwardConcentrating on value creation for customers, the centre adopted knowledge management and a design-for-quality methodology concurrently, was designated a key engine of the group's businesses, focused on selected technological areas, and upgraded the R&D process from one of planning to one of commercialisation.
The third period is the one to notice. A move toward long-term basic research — the textbook remedy for a corporate lab captured by its divisions — made the centre more efficient and no better at its actual mission. What the paper credits with the turnaround is the opposite manoeuvre: re-basing the lab on business-unit alignment.
The four-direction transformation model
This is the core teachable framework. Four directions, run concurrently rather than in sequence, each with concrete rules attached.
The four directions
Visioning and restructuring
Assess existing projects and core technology areas first; the diagnosis was that initial research areas and capabilities were "hardly adequate" for the mission. Rearrange all R&D projects around a small set of core technology areas, selected on one criterion — that the area is closely linked both to existing businesses and to new business opportunities. Five were chosen: digital technology, nanotechnology, optoelectronics, energy and biotechnology.
Excellence of product
Every project targets "the first or the best" — a fundamental patent or a state-of-the-art technology. At planning, assess the technological development status of leading global organisations on the key technologies, then set project goals above that benchmark. Build relationships with divisions by sharing business strategies and technology roadmaps; any project intended for near-term commercialisation must specify the business strategy it serves and the roles responsible for value creation.
Excellence of process
Customise one quality methodology into the whole R&D process, from planning through gates to commercialisation, so that the centre, development and manufacturing share documentation forms and terminology. Run gates as design reviews. Start technology transfer before the project finishes. Use technology-tree analysis to set project priority and structured-inventive-thinking methods to explore alternative solution paths.
Excellence of people
A matrix of technology groups against projects: the groups build core capability around key technologies and spot emerging opportunities; project managers develop the specified activities and commercialise the results. Recruit globally rather than only from domestic graduates. Operate a career-level system used for three purposes — forming project teams, recruiting, and career planning — with project leadership gated on level. Evaluate on two axes: project results, and potential capability in key technology areas.
Three of those rules are unusual enough to be worth lifting on their own, whatever you make of the case. First, the gate is repurposed. Traditional stage-gate management reviews progress and makes a proceed decision; here the emphasis falls on discussion between technology experts and business professionals about critical issues and potential problems. It is a knowledge-sharing forum that happens to have a decision attached, which is close to the argument made for structured learning events in post-project reviews in R&D.
Second, transfer starts before completion. Project teams collaborate with business divisions while the project is still running, so commercialisation problems surface while there is still runway to solve them. Third, project leadership is gated on an explicit career level rather than on availability — only researchers above a defined level may lead.
Where each tool sits in the process
The paper's single figure is a two-stage process spine with tool bands underneath, each band spanning only the phases where that tool applies. The reader's take-away is the sequencing, not the tool list.
- Idea generation
- World-class goal setting
- Milestones and planning
- Research and development
- Technology transfer
- Business unit as customer
TOOL BANDS AGAINST THE PROCESS SPINE, AS DRAWN IN R11
| Tool | Where it spans | What it is used for |
|---|---|---|
| Value curve | From idea generation | Analysing diverse customer needs at the planning stage, alongside quality function deployment |
| Technology roadmap | From idea generation across the whole initiation stage | Aligning the project with the business unit's strategy; coordinating roles, resources and commercialisation plans |
| Technology tree and structured inventive thinking | Around goal setting and milestones and planning | Identifying key technologies and relationships among core technologies; predicting evolution trajectories to find alternative solutions |
| Design of experiments and robust design | From planning into research and development | Statistical and robustness work in the middle of the project |
| Quality management | Across technology transfer and the customer end | Governing the handover |
| Customer feedback | Running back from the customer end to the front | Closing the loop from the business unit to idea generation |
Reconstructed from the single figure in one case study. The spans are as drawn in the paper; the paper reports no per-tool outcome data.
Two things follow. The R&D lifecycle terminates not at technology transfer but at the business unit as customer, with a closed feedback loop back to idea generation. And each tool is bound to a phase rather than applied uniformly — creativity and positioning tools at the front, statistical and robustness tools in the middle, quality management at the handover.
What the paper says drove it
- Consensus on the need for organisational change
- Strong leadership by the top-management team
- Close alignment with business units
- Stable corporate investment
- Actionable planning and performance management
- Above all, a clear strategic direction as a corporate R&D centre, and close relationships with business divisions from planning through ongoing management to commercialisation
The mechanism claimed is integration: research methodologies such as technology roadmapping and scenario management were combined with development methodologies such as robust stage-gate processes, and every project was aligned at initiation to the business and technology strategies of the business units through shared roadmaps. That alignment is described as a key success factor and as the thing that clarified roles and responsibilities in commercialisation.
Evidenced and asserted — the split, kept explicit
R11 mixes counted quantities with causal claims in the same paragraphs. Separating them is not a scholarly nicety here; it is the difference between a defensible reference and an unusable one.
WHAT R11 EVIDENCES AND WHAT IT ASSERTS
| Claim | Status | Qualification |
|---|---|---|
| Commercialisation rate of 18 percent (1997), 61 percent (2002) and 80 percent (2004) | Evidenced as measurement | Internally corroborated by the underlying cohort counts. No source, method or definition of "reached commercialization" is stated anywhere in the paper |
| Of 133 projects conducted up to 1997, 24 were transferred to business units and contributed commercial value; the other 109 contributed hardly at all | Evidenced as counts | The split of the 109 is internally inconsistent as printed — see the note below |
| Of 79 recently completed projects, 33 were transferred successfully and 15 were in the process of commercialisation | Evidenced as counts | Not commensurable with the 1997 figure: 133 is a cumulative census across roughly a decade, 79 is a cohort of unstated window |
| Applications filed outside the home jurisdiction moved from 19 percent of 380 applications in 1997 to 85 percent of 1,400 in 2004 | Evidenced as counts | Numerator and denominator both change, so volume growth and strategy change are conflated in the percentage |
| Approximately 1,000 researchers, 86 percent holding graduate degrees | Evidenced as a stated figure | Single point in time, no trend |
| That the transformation caused the performance improvement | Asserted | No counterfactual, no control, no statistical test, and no attempt to separate the transformation from the group's own growth, from market conditions, or from the 1993 group-wide philosophy change |
| That business-unit satisfaction with R&D results "grew rapidly" | Asserted | No survey, no instrument, no scores, no respondents |
| The six key drivers of the transformation | Asserted | No comparative case, no variance analysis, no test of any driver's necessity or sufficiency |
| The efficacy of any individual tool — technology tree, structured inventive thinking, value curve, quality function deployment, design review | Asserted | No per-tool outcome data; the sidebar is narrative |
| That world-class goal setting evoked research challenges and motivation | Asserted | No motivation or engagement measure |
| That the transformation was painful and some researchers left | Asserted | No attrition rate, no headcount before or after |
All counted figures are findings of one organisation's records as reported in one case study. None is a benchmark.
The group's own sales grew steeply across the same period, which the paper partly attributes to market conditions it does not control for. Any reading that treats the centre's commercialisation rate as the effect of the transformation has to account for that separately, and the paper does not.
Using a single case honestly
A case study earns its place by showing a mechanism in enough detail to be copied and tested, not by proving that the mechanism works. Read R11 that way and it is a strong source; read it as evidence of effect and it will not hold.
Before you cite this case in a business case
- State that it is one organisation, one direction of travel, written by insiders, with no method section.
- Quote the commercialisation figures as counts from that organisation's records, never as a benchmark or an expected result.
- Do not carry the causal claim across. Carry the mechanism — alignment through shared roadmaps, design reviews, early transfer, level-gated leadership — and test it yourself.
- Say which of the four directions you are actually adopting. Adopting the tool list without the structural move is adopting the least evidenced part.
- Name the confound in your own setting, as the paper does not: what else changed at the same time, and how would you tell the difference?
For the analytical apparatus that sits underneath decisions of this kind — evaluation tools, portfolio displays and the organisational conditions for making them stick — see R&D project evaluation tools and governing R&D decisions organisationally. For a multi-firm empirical treatment of the same question this case answers narratively, see what distinguishes successful R&D projects. For how a case study compares with the other kinds of paper in this set, see eleven R&D management papers compared.
What to carry forward
- The four-direction model — visioning and restructuring, excellence of product, process and people — is the transferable content, and it was run concurrently rather than in sequence.
- The straddle between research centre and development centre is diagnosed as the problem to resolve, not the balance to preserve.
- Gates as design reviews, transfer before completion, and leadership gated on career level are the three rules most worth lifting.
- R11 evidences that its numbers changed. It asserts that the transformation caused the change, with no counterfactual and no control.
- The two cohorts behind the commercialisation rate are not comparable with one another, and the paper's split of 109 projects does not close as printed.
- A single insider case study is a source of mechanism, not a source of effect size.
Frequently asked questions
Can I use the 18 to 80 percent commercialisation improvement as a target?
No. Those are counts from one organisation's records, reported in a case study with no stated method and no definition of what "reached commercialization" means. The two cohorts behind them are also not comparable with each other — one is a cumulative census to 1997, the other a recent cohort of unstated window. Treat them as description of one case, not as an achievable rate.
What is the single most useful idea in R11?
That a corporate lab straddling basic research and applied development, serving divisions at very different capability levels, is diagnosed as dysfunctional rather than balanced. The transformation resolves the straddle by aligning every project to a business unit's strategy through shared technology roadmaps. That is a structural move, and it is the part the tool list depends on.
How does a design review differ from a stage gate?
In R11 the gate keeps its position in the process but changes its purpose. Instead of primarily reviewing progress and issuing a proceed decision, it convenes technology experts and business professionals to discuss critical issues and potential problems. The paper treats it as a cross-boundary knowledge-sharing forum, which is why it pairs naturally with transferring technology before the project finishes.
Does the paper show that its quality methodology worked?
It reports that the methodology was adopted and that performance figures improved, but it offers no per-tool outcome data and no comparison group. It also concedes in its closing section that the methodology may be insufficient for genuinely new research fields, and that development-oriented and research-oriented projects need different treatments. The efficacy claim is asserted, not evidenced.
Why does the arithmetic in the project counts matter?
Because it tells you how much scrutiny the numbers received before publication. The split of 109 non-contributing projects only coheres if a word printed as "successful" means the opposite, and the 79-project cohort's date has to be inferred from a percentage rather than read off the page. Neither invalidates the figures; both are reasons to quote them as reported counts with the ambiguity attached.
Is this model only relevant to technology followers?
The paper says so itself. It argues that existing prescriptions derive from the experiences of technology leaders and positions itself as filling the gap for followers. That framing is honest, and it cuts both ways — the case is not evidence about what a leading laboratory should do, and the paper names the shift from follower to leader as a challenge that will require the model to change.
References and source attribution
- R11 — how a large electronics manufacturer transformed its corporate R&D centre. Single-organisation descriptive case study in a practitioner journal for R&D and technology managers, 2006; 6 printed pages; one figure and one sidebar worked example; six book references. No stated research method, no description of data collection, sampling or analysis. One author had previously worked in the studied organisation's chief technology office and the other was a serving executive of the parent group. Sections used here: the four-period evolution, the four-direction transformation model, the key drivers, the sidebar example, and the separation of evidenced from asserted outcomes.
- The six works cited within R11 — books on disruptive innovation, innovation S-curves, the dynamics of innovation, corporate research laboratories, and third- and fourth-generation R&D — are invoked only for framing propositions. No quantified claim is imported from any of them, and none was supplied to this library.
- Eleven copyrighted journal articles on R&D project management, supplied as a reading set assembled by a student for a literature review and profiled for this library. Front matter, abstracts, framework sections, tables and figures were read; article bodies were not reproduced, and all content here is paraphrase. R11 is the only case study in that set, and the set is not a systematic or representative survey of the field.
- Supplied teaching source for this library (research methods and research process materials). Used here for page conventions, voice, and the treatment of case-study evidence; it does not treat corporate R&D organisation.
Suggested questions for Ask KEVOS
- Turn the four-direction transformation model into a diagnostic for our own R&D function.
- Draft a design review agenda that convenes technical and business participants rather than issuing a proceed decision.
- What would we have to measure to test the causal claim R11 only asserts?
- How would we select core technology areas using the linkage test from direction one?
- Write the limitations paragraph for citing a single insider case study in a business case.
- Compare transferring technology before project completion against a conventional handover at closure.