Horizon Scanning is an exploratory thinking method for identifying early signals of changes that may have important future consequences. It involves gathering information from a wide range of fields, including technology, society, economics, politics, and the environment. It is one of the principal methods of Strategic Foresight and is used in public policy, corporate strategy, research and development, new business development, and risk management.
When organizations examine their future business environments, they typically collect information about market trends, technological developments, competitors, and customer needs. However, conventional market and technology research tends to focus on subjects that are already recognized as important or that belong to established categories.
Major changes do not necessarily begin as visible developments in market statistics or the activities of leading companies. Small-scale experiments with new technologies, unusual applications, changes in customer behavior, and emerging regulatory proposals may eventually contribute to transformations in entire industries.
Horizon Scanning focuses on these early indications of change.
For example, when examining the future adoption of AI in manufacturing, it is not sufficient to investigate only AI market growth rates or investments by major companies. Small-scale experiments in manufacturing facilities, unconventional applications, changes in technology costs, and regulatory developments affecting industrial data may also provide important clues.
These observations can then be compared and interpreted to examine what kinds of future changes they might indicate.
The purpose of Horizon Scanning is not to predict the future accurately, but to discover emerging signals of change and enable organizations to consider potential opportunities and risks before they become widely recognized.
However, merely collecting unusual news does not constitute effective Horizon Scanning. It is essential to verify information, examine relationships among different signals, and clarify their implications for decision-making.
Origins and Development of Horizon Scanning
Approaches to recognizing future changes at an early stage have developed through strategic management, military planning, and futures studies.
One influential concept is that of Weak Signals.
In his 1975 paper “Managing Strategic Surprise by Response to Weak Signals,” strategic management researcher H. Igor Ansoff emphasized the importance of recognizing possible environmental changes from incomplete and ambiguous early information and preparing responses before those changes become fully apparent.
When organizations finally recognize a major environmental change, they may already lack sufficient time to respond effectively.
Recognizing early indications makes it possible to begin gathering information and preparing responses without waiting until every aspect of the emerging development becomes certain.
Horizon Scanning subsequently became established as a component of Strategic Foresight and was further developed by government agencies and research institutions.
The UK Government Office for Science describes Horizon Scanning in The Futures Toolkit as a systematic approach to collecting and examining emerging developments and signals that may influence future policy and strategy.
The OECD and European Commission also recognize Horizon Scanning as an important approach to identifying technological and social changes and incorporating them into policy and strategic planning.
More recently, traditional literature reviews and expert consultations have been supplemented by online information collection, text mining, machine learning, and Large Language Models (LLMs).
However, automating information collection is not equivalent to understanding future change.
Effective Horizon Scanning requires both discovering information and interpreting its possible significance.
What Are Weak Signals?
A weak signal is an early indication of a potentially important future change whose meaning, implications, or eventual impact are not yet clear.
For example, when a technology has already been widely adopted in the principal products of major companies, recognizing its importance is relatively straightforward.
However, a technology that is being tested only at a research institution, or a new business model being tried by a small number of companies, may be much harder to assess.
At that stage, it may be unclear whether the development will eventually have a substantial impact.
Horizon Scanning seeks to examine what such early developments might mean.
However, information should not be considered an important weak signal merely because it is scarce, unusual, or new.
Many new technologies and business experiments never achieve widespread adoption. Technologies initially regarded as promising may also be replaced by alternative approaches.
It is therefore important to distinguish observed facts from interpretations and hypotheses about the future.
For example:
| Item | Description |
|---|---|
| Observed fact | A company has begun testing a new AI monitoring method on specialized manufacturing equipment |
| Interpretation as a signal | An alternative approach to conventional equipment management may be emerging |
| Future hypothesis | If the method becomes inexpensive, it could change how existing equipment is managed |
| Required verification | Technical performance, implementation cost, applicability to other equipment, customer demand |
This is a hypothetical example intended to illustrate the method.
The important distinction is between what has actually been observed and what the observation might imply.
A single observation may support several different hypotheses about future developments.
Signals, Trends, Emerging Issues, and Drivers of Change
Horizon Scanning uses concepts such as Signals, Trends, Emerging Issues, and Drivers of Change.
These concepts are related but are not interchangeable.
| Concept | Meaning | Example |
|---|---|---|
| Signal | An observable event or piece of information suggesting possible change | A factory begins testing a new AI-based maintenance method |
| Weak Signal | An early indication whose future significance remains uncertain | A small-scale experiment with an unconventional equipment management approach |
| Trend | An identifiable direction of change over time | Increasing adoption of AI monitoring systems in manufacturing |
| Emerging Issue | A developing question or concern that may become important for organizations or society | Responsibility and safety requirements for AI-managed industrial equipment |
| Driver of Change | A factor that promotes, constrains, or shapes broader developments | Labor shortages, declining AI costs, regulatory changes |
These concepts do not necessarily represent stages in a fixed sequence.
Several signals may contribute to the recognition of a trend, while established drivers of change may give rise to new issues.
A Megatrend refers to a broad, long-term development that affects society or the economy on a substantial scale, such as demographic transformation or long-term urbanization.
Horizon Scanning examines not only established megatrends but also emerging signals that may alter their direction or significance.
Events that would have major consequences if they occurred, but whose likelihood is low or difficult to estimate, are sometimes considered as Wild Cards.
A Wild Card is a hypothetical future event and should be distinguished from a signal that has already been observed.
The Difference Between Trend Analysis and Horizon Scanning
Trend Analysis examines historical and current data to identify the direction and rate of change.
For example, researchers may analyze changes over time in the number of scientific papers, patent applications, products, or market transactions associated with a particular technology.
Increasing activity may indicate growing research interest or market attention.
However, increasing numbers do not necessarily mean that the technology will succeed.
Nor do important technological transformations always appear immediately as identifiable statistical trends.
Horizon Scanning also examines small-scale developments, fragmented information, and phenomena that deviate from established trends.
| Aspect | Trend Analysis | Horizon Scanning |
|---|---|---|
| Main focus | Identifiable patterns of change | Signals that may become important in the future |
| Information characteristics | Often supported by continuous datasets | May involve scarce or fragmented information |
| Central question | What is changing, and how quickly? | What changes may be emerging but remain insufficiently recognized? |
| Typical methods | Time-series analysis, statistics, indicator comparison | Broad exploration, signal assessment, interpretation |
| Main results | Direction and magnitude of change | Emerging opportunities, risks, and subjects requiring further investigation |
The two approaches are complementary.
Continued monitoring of a signal discovered through Horizon Scanning may eventually reveal a measurable trend.
Conversely, Trend Analysis may identify an unexpected deviation that deserves further investigation as a possible signal of structural change.
Defining the Scope of Exploration
An effective Horizon Scanning activity begins by clarifying the purpose of the investigation.
For example, a manufacturing company might ask:
“What developments could change demand for our core products or transform our competitive environment over the next ten years?”
An R&D department might instead ask:
“What technologies or phenomena outside our current research domains could become important in the future?”
However, defining the scope too narrowly can cause unexpected developments to be overlooked.
For example, if an industrial measurement equipment manufacturer monitors only news from the measurement equipment industry, it may miss developments in software business models, semiconductor cost reductions, or workforce changes that could alter its future competitive environment.
Horizon Scanning can therefore combine two forms of exploration.
The first focuses on developments directly relevant to a particular strategic decision.
The second deliberately explores a broader range of fields beyond the organization’s existing business domain.
The first helps maintain decision relevance.
The second increases the possibility of discovering changes that were not anticipated within the original assumptions.
The purpose of exploration should be clearly defined, but the subjects being explored should not be restricted exclusively to familiar industries or keywords.
Organizing Exploration Areas with STEEP and PESTLE
Frameworks such as STEEP and PESTLE are often used in Horizon Scanning to reduce the risk of overlooking important domains.
STEEP represents five broad areas: Social, Technological, Economic, Environmental, and Political.
PESTLE distinguishes Political, Economic, Social, Technological, Legal, and Environmental factors.
When examining the future of manufacturing, the following areas may be considered.
| Domain | Examples of Information to Explore |
|---|---|
| Social | Labor shortages, skill transfer, working practices, changing customer values |
| Technological | AI, sensors, materials, manufacturing methods, communication technologies |
| Economic | Capital investment, energy prices, supply chains, international competition |
| Environmental | Decarbonization, resource use, environmental regulations, climate change |
| Political and Institutional | Industrial policy, trade policy, national security, regulations, standards |
These frameworks help reduce gaps in information collection.
However, actual developments rarely belong exclusively to one category.
For example, lower AI costs may influence capital investment, workforce requirements, product prices, regulation, and competition.
Information that does not fit neatly into an established category should not be forced into one unnecessarily.
STEEP and PESTLE are tools for checking the breadth of exploration, not for confining future developments within predefined categories.
The Basic Horizon Scanning Process
Drawing on approaches described in resources such as the UK Government’s Futures Toolkit, Horizon Scanning can be organized as a continuous process extending from information collection to strategic interpretation.
- Define the exploration objective. Clarify which decisions the activity should support, the relevant domains, and the time horizon.
- Design the information sources. Select diverse sources, including technical literature, news, patents, policy documents, corporate activities, and expert knowledge.
- Collect signals of change. Record observations that may indicate potentially important future developments.
- Verify the information. Confirm sources, dates, and factual details, distinguishing observations from assumptions and interpretations.
- Organize and connect signals. Group related observations and examine patterns and relationships that may indicate broader changes.
- Evaluate significance and uncertainty. Consider potential impact, relevance, plausibility, and evidence quality.
- Examine strategic implications. Identify potential opportunities, threats, questions requiring further investigation, and possible strategic responses.
- Continue monitoring. Incorporate new information and revise assessments and future hypotheses.
The process is not necessarily linear.
Additional research may reveal that an initial interpretation was incorrect.
Alternatively, new relationships among signals may produce entirely different questions for investigation.
Horizon Scanning is therefore better understood as an ongoing process for updating an organization’s understanding of change than as a one-time report.
How Should Signals Be Recorded?
In Horizon Scanning, observations should be recorded in a form that allows subsequent comparison and verification.
One practical approach is to create a Signal Card for each potentially important development.
| Field | Description |
|---|---|
| Signal title | A concise name |
| Observed fact | What actually happened |
| Event and publication dates | When the event occurred and when the information became available |
| Information source | Scientific paper, patent, corporate announcement, news article, field observation, etc. |
| Related domains | Technology, society, economics, institutions, etc. |
| Interpretation | What may be beginning to change |
| Potential future impact | Possible opportunities or risks |
| Evidence quality | Reliability of sources and verification status |
| Questions for investigation | What remains unknown |
| Update history | New information and changes in interpretation |
This structure helps distinguish future-oriented hypotheses from the evidence on which they are based.
It also allows analysts to examine how their initial assessments change as additional information becomes available.
When LLMs are used, it is particularly important to distinguish AI-generated interpretations from information verified through original sources.
Evaluating and Prioritizing Signals
It is rarely practical to investigate every collected signal in equal depth.
Some form of evaluation is therefore necessary to allocate limited analytical resources.
Possible evaluation criteria include:
| Criterion | Key Question |
|---|---|
| Impact | How significant could the change become if it spreads? |
| Novelty | How does the development differ from established technologies, business models, or trends? |
| Relevance | How might it affect the organization’s business, R&D, or policy objectives? |
| Plausibility | Is there a credible mechanism through which the proposed change could occur? |
| Evidence Quality | How reliably can the underlying observations be verified? |
| Time Horizon | When might the change become significant? |
| Uncertainty | What remains unknown about its future development? |
These assessments can be represented numerically when appropriate.
For example, impact and relevance might be evaluated using five-point scales to help identify signals requiring further investigation.
However, simply combining all criteria into a weighted total can be problematic.
Early-stage signals often have limited supporting information.
If weak evidence automatically leads to a low total score, the method may eliminate precisely the emerging developments that Horizon Scanning is intended to discover.
A signal may have potentially substantial consequences while remaining highly uncertain.
Conversely, a clearly documented development may have little strategic significance.
Instead of forcing every signal into a single ranking, organizations may classify them into categories such as:
- High strategic priority.
- Requires further investigation.
- Continue monitoring.
- Currently low priority.
The purpose of signal evaluation is not to determine the probability of future success, but to decide where limited research and monitoring resources should be allocated.
Example: Scanning the Future of AI in Manufacturing
Consider an industrial measurement equipment manufacturer exploring how AI and data technologies might change its future business environment.
The company traditionally develops and sells high-precision measurement equipment.
It now wants to understand how its products and customers’ equipment management practices might evolve.
The following signals are hypothetical examples illustrating the analytical method, not findings from an actual market study.
| Signal | Example of an Observed Development | Possible Future Implication |
|---|---|---|
| A | Low-cost edge AI is tested for condition monitoring in small manufacturing facilities | Equipment monitoring may become less dependent on expensive dedicated systems |
| B | Software is developed to integrate data from existing sensors | Greater emphasis may be placed on using existing data rather than replacing equipment |
| C | Maintenance contracts linked to equipment operating conditions are tested | Equipment manufacturers may shift toward operational support services |
| D | Restrictions on transferring industrial data outside facilities become stricter | On-premises and edge-based analytics may become more valuable |
Each signal can be evaluated separately.
However, examining their relationships may reveal broader implications.
If A and B develop together, manufacturers may become able to analyze existing equipment data at relatively low cost.
If C also develops, equipment manufacturers may find opportunities to expand into monitoring and maintenance services.
On the other hand, if D becomes important, locally deployed analytical systems may be more attractive than cloud-centered services.
The objective is not to conclude that these developments will occur.
It is to examine the conditions under which several signals could combine to create different technological architectures or business structures.
Discovering Patterns Behind Individual Signals
Collecting large numbers of signals is not sufficient if they remain isolated news items.
Important changes may become visible only when relationships among observations are examined.
Consider three hypothetical developments from different industries:
- In medical devices, methods are being investigated for analyzing measurement data directly within the device.
- In manufacturing equipment, low-power local devices are being explored for anomaly detection.
- In logistics, systems are being used to extract only essential information at the device level to reduce communication requirements.
These developments belong to different industrial domains.
When classified by product or industry, they may appear unrelated.
However, examining their functions and structures may reveal a common pattern:
Rather than transmitting all data to a centralized system, information is selectively processed near its source.
Recognizing this common structure may suggest a broader technological change that is not immediately apparent from individual observations.
However, several similar examples do not automatically establish a new trend.
The technical conditions and reasons for adoption must be investigated to determine whether the developments genuinely share a common mechanism.
At this stage, comparing functions, structures, characteristics, and relationships can be more useful than relying only on matching terminology.
The Number of Signals Is Not the Same as Their Importance
Text mining and statistical analysis can measure how frequently particular terms appear and whether their frequency increases over time.
For example, rapid increases in scientific papers, patent applications, or news articles concerning a technology may indicate growing research activity or public attention.
However, frequency does not necessarily indicate future importance.
A widely discussed technology may generate large numbers of articles repeating essentially the same information.
Meanwhile, an important early-stage development may be recognized by only a small group of researchers or organizations.
If signals are ranked primarily by information volume, well-established topics may receive high scores while unfamiliar developments are overlooked.
Keyword frequency and growth rates can provide useful clues, but they are not sufficient measures of the significance of weak signals.
It is also necessary to examine whether a small amount of information contains an unfamiliar function, structure, relationship, or business implication.
This is one of the central challenges of Horizon Scanning.
Relationship with Scenario Analysis
Horizon Scanning and Scenario Analysis are complementary methods within Strategic Foresight.
Horizon Scanning identifies early indications of future change from currently observable information.
Scenario Analysis examines combinations of important uncertainties to construct different possible future environments.
For example, Horizon Scanning may identify signals involving declining AI costs, changes in industrial investment, and new restrictions on data use.
Examining these developments may suggest that the level of capital investment and the speed of AI adoption are particularly important uncertainties.
Scenario Analysis can then use these uncertainties to construct several alternative futures and evaluate how strategies perform under each.
| Aspect | Horizon Scanning | Scenario Analysis |
|---|---|---|
| Main objective | Identify early signals of change | Evaluate strategies under alternative futures |
| Starting point | Currently observable information | Important drivers and uncertainties |
| Main activities | Exploration, collection, comparison, interpretation | Scenario construction, strategic comparison, evaluation |
| Main results | Signals, drivers of change, questions for investigation | Scenarios, strategic assessments, response options |
Horizon Scanning can provide inputs for constructing scenarios.
Conversely, Scenario Analysis can identify important uncertainties that should be monitored through continuing Horizon Scanning.
Combining the two methods creates a process in which future possibilities are examined and strategic assumptions are updated as new information emerges.
Relationship with SWOT and TOWS Analysis
SWOT Analysis organizes internal strengths and weaknesses and external opportunities and threats.
However, the opportunities and threats included in SWOT depend on what the analysts already know.
If the analysis concentrates exclusively on current market conditions, it may overlook technological or institutional changes that could transform the future competitive environment.
Horizon Scanning can identify emerging developments that suggest previously unrecognized opportunities or threats.
These can then be incorporated into SWOT and used to generate new strategies through TOWS Analysis.
However, a weak signal does not necessarily represent an established opportunity or threat.
Its actual significance may remain uncertain.
Rather than immediately classifying every signal as a confirmed opportunity or threat, it may be more appropriate to record it as a hypothesis requiring investigation.
Relationship with Concept Exploration
Horizon Scanning shares an important concern with Concept Exploration, an approach proposed by Mindware Research Institute.
One of the original motivations behind Concept Exploration was to observe technological and social developments continuously and discover newly emerging concepts.
In this respect, it overlaps with Horizon Scanning’s interest in identifying early signals of change.
However, the two approaches do not have exactly the same emphasis.
Horizon Scanning focuses on collecting signals that may indicate future developments and evaluating their possible implications.
Concept Exploration also emphasizes identifying common characteristics, structures, and relationships among individual observations and recognizing them as emerging concepts.
For example, Horizon Scanning may collect signals concerning similar technological developments occurring in several industries.
Concept Exploration may then examine the common functions and structures underlying those developments and abstract them into a new conceptual framework.
Concept Exploration also includes another direction: searching for unknown objects or technologies based on previously recognized characteristics and structures, without requiring advance knowledge of their names or domains.
This direction subsequently developed into Cross-Domain Concept Exploration.
The relationship can be summarized as follows.
| Aspect | Horizon Scanning | Concept Exploration |
|---|---|---|
| Primary concern | Early indications of potentially important future changes | Formation of new concepts and discovery of unknown possibilities |
| Starting point | Observable signals and developments | Observations, characteristics, structures, patterns |
| Central question | What is beginning to change? | What new concepts are emerging, and what should we explore? |
| Direction of exploration | From signals toward possible future changes | From information to concepts, and from concepts to unknown objects |
| Main results | Signals, drivers of change, emerging issues | New concepts, unknown objects, exploration hypotheses |
In practice, the two approaches may form part of a single process.
Signals collected through Horizon Scanning may lead to the recognition of a new concept.
That concept may then guide further investigation into unfamiliar technological domains.
If Horizon Scanning asks, “What is beginning to change?”, Concept Exploration asks, “How can these changes be understood as emerging concepts, and what unknown possibilities can those concepts help us discover?”
Horizon Scanning with AI
LLMs and machine learning are changing the practical methods available for Horizon Scanning.
Traditionally, signal collection required researchers to review news, scientific literature, policy documents, and corporate information continuously.
AI can now assist in collecting documents from information sources, summarizing their contents, classifying them, and extracting statements that may indicate emerging developments.
For example, AI can extract new functions or applications described in technical literature and compare them with previously collected information.
It can also help identify similarities and common themes among signals originating from different sources.
However, simply automating keyword searches does not overcome the limitations of conventional research.
If the search remains restricted to familiar technologies and market categories, AI may still overlook important developments occurring outside those domains.
LLM-generated summaries and significance assessments also require caution.
LLMs may describe established topics effectively while struggling to evaluate highly localized, unfamiliar, or very recent developments for which little information is available.
Singapore’s Centre for Strategic Futures, in a 2026 account of its Horizon Scanning practices, also discussed the benefits of AI-assisted information processing while highlighting difficulties associated with discovering new and localized signals.
Combining AI-assisted collection with human exploration and interpretation therefore remains important.
Discovering Change Through Text Mining
Horizon Scanning can extend beyond measuring keyword frequencies or the number of documents associated with a topic.
It may also be useful to examine changes in the semantic structure of information.
For example, technical documents can be converted into embedding vectors and organized according to semantic similarity.
This can help visualize relationships among technological topics across different domains.
Methods such as Growing Neural Gas (GNG) can be used to construct models of conceptual structure, helping researchers explore clusters of similar information and relationships among them.
Comparing structures constructed from information collected at different times may reveal newly emerging groups of topics or changes in relationships among existing groups.
However, such analysis presents several challenges.
If the coverage of information sources changes, the document distribution may change even when the underlying technological environment has not.
Likewise, changing the embedding model, dimensionality reduction method, or clustering parameters may alter the apparent conceptual structure.
Longitudinal analysis should therefore maintain consistent analytical conditions where possible and distinguish genuine changes in the data from changes introduced by the model.
Furthermore, the appearance of a new cluster does not automatically indicate an important future development.
Machine learning can identify candidates for further investigation, but it cannot independently establish their strategic significance.
Limitations of Horizon Scanning
Horizon Scanning has several fundamental limitations.
First, it depends on the available information sources. If those sources are concentrated in particular industries, languages, regions, or academic disciplines, the results may reflect those biases.
Second, distinguishing important signals from merely unusual events is difficult. At an early stage, there may be insufficient evidence to evaluate future consequences.
Third, analysts’ experience and expertise influence interpretation. They may favor information consistent with their existing expectations and underestimate developments that challenge those expectations.
Fourth, information collection may become an end in itself. Accumulating large numbers of signals has limited practical value unless they contribute to decisions.
Fifth, the meaning of signals can change over time. Technologies initially considered important may fail to spread, while overlooked technologies may develop rapidly.
Sixth, hindsight can distort assessments of past developments. Looking backward, it may appear obvious that important signals were already present.
However, many other signals may have appeared equally promising at the time.
Recording the information and judgments available when a signal was first identified can help reduce this hindsight bias and support later evaluation.
Horizon Scanning is therefore better understood as a method for continuously observing, interpreting, and updating understanding under uncertainty than as a technology for predicting the future with certainty.
Horizon Scanning in ThinkNavi
ThinkNavi regards Horizon Scanning as a thinking-support method for collecting early signals of potentially important future developments and examining their meaning over time.
ThinkNavi Free could provide a framework in which users define exploration themes and record signals using structured cards.
Each card could contain observed facts, sources, dates, related domains, interpretations, potential future implications, and questions requiring further investigation.
Users could also evaluate signals according to potential impact, novelty, relevance, plausibility, and evidence quality.
However, these evaluations should not necessarily be reduced to a single total score.
It would be useful to distinguish the potential significance of a signal from uncertainty about the supporting information.
For example, a signal that could have substantial future consequences but currently lacks sufficient evidence should not automatically receive a low priority.
It may instead be identified as requiring further investigation.
Users could also establish relationships among multiple signals and organize them into broader themes of change.
As new signals are added, existing themes could be revised or reorganized.
Integration with Scenario Analysis would allow important signals to inform the identification of future drivers and uncertainties.
Integration with SWOT and TOWS Analysis would allow emerging opportunities and threats to contribute to strategic planning.
An LLM-assisted version could support extracting candidate signals from information sources, summarizing observations, classifying developments, identifying related signals, proposing further research questions, and generating hypotheses about possible future implications.
However, AI-generated interpretations should be distinguished from verified source information, and the evidence supporting each signal should remain traceable.
Further integration with the ConceptMiner product family could support organizing signals through embeddings and analyzing their conceptual structure using GNG and related techniques.
This could help users investigate not only where new signals are appearing, but also whether common functions or structures are emerging across otherwise unrelated fields.
Patterns identified through such analysis could also guide additional exploration beyond familiar keywords and technological domains.
These possibilities suggest an integrated approach in which Horizon Scanning collects signals, Concept Exploration identifies emerging concepts and common patterns, and Cross-Domain Concept Exploration investigates unknown objects and applications sharing those patterns.
However, the principal value does not lie in collecting the largest number of documents or constructing the most elaborate analytical model.
The significance of Horizon Scanning lies in recognizing changes that are not yet widely understood and reconsidering assumptions about the future that have previously been taken for granted.
Achieving this requires more than efficiently collecting familiar information. It also requires attention to developments that are difficult to explain within established categories and to changes that share underlying structures across different fields.
AI and machine learning can support this process, but human interpretation and judgment remain essential for understanding the meaning of emerging signals, examining possible futures, and translating insights into action.
Primary English Term: Horizon Scanning
Related Terms: Weak Signals, Emerging Issues, Drivers of Change, Trend Analysis, Strategic Foresight, Environmental Scanning
Related Thinking Methods:
- Concept Exploration | Discovering Emerging Concepts and Unknown Possibilities
- Cross-Domain Concept Exploration,
- STEEP and PESTLE Analysis,
- Scenario Analysis | Exploring Multiple Futures and Developing Strategies Under Uncertainty
- SWOT and TOWS Analysis | From Environmental Assessment to Strategic Hypothesis Generation
- Systems Thinking — Understanding Interdependence, Feedback, and Change Over Time
- Cross-Impact Analysis,
- Text Mining,
- Structure-Mapping Theory | Understanding Relational Correspondences and Generating New Inferences Across Domains
- Sensitivity Analysis | Examining How Changes in Assumptions Affect Model Results
References and Further Reading
- H. Igor Ansoff, “Managing Strategic Surprise by Response to Weak Signals,” California Management Review, Vol. 18, No. 2, pp. 21–33, 1975. Paper
- UK Government Office for Science, The Futures Toolkit. Practical guidance on futures methods, including Horizon Scanning.
- UK Government Office for Science, Horizon Scanning: Template, 2024. A template for recording and assessing signals.
- OECD, Strategic Foresight. An overview of approaches including Horizon Scanning, megatrends, and Scenario Analysis.
- OECD, Tools for Agility: Actionable Strategic Intelligence and Policy Experimentation, OECD Science, Technology and Innovation Outlook 2025.
- European Commission, Strategic Foresight.
- United Nations Development Programme, Horizon Scanning.
- Centre for Strategic Futures, Singapore, Honing Our Edge: A Peek into CSF’s Horizon Scanning Practice, 2026.
- Government of Canada, Policy Horizons Canada, Module 3: Scanning.