01 Case Study · Research

Stakeholder Intelligence

Using AI and natural language processing to close the communication gap in multi-stakeholder project governance.

MSc Engineering Management dissertation, Anglia Ruskin University. Supervised by Dr. Ana Cocho Bermejo.

Survey group21 project & IT leaders
Technical stackNLP · ML · Power BI
Headline finding81% agreed AI aids communication
Year2024
02 Method

From raw data to strategic intelligence

The framework moves project communication through four stages, turning unstructured discussion into governance action.

Raw communicationMeeting data, RAID logs
NLP analysisSentiment & tone detection
AI synthesisRisk identification
Strategic actionPM conflict resolution
03 Synergy

Human and AI collaboration

Project failure is rarely a technical issue; it is a communication breakdown. The framework positions AI as a force multiplier that automates synthesis, freeing the project professional to lead with high-value strategic mediation.

Governance domainAI-automated efficiencyHuman-in-the-loop strategy
ReportingReal-time status & synthesisStakeholder contextualisation
Risk managementLinguistic sentiment flaggingStrategic mediation & mitigation
Resource planningPredictive bottleneck forecastingStrategic talent & skill allocation
04 Evidence

What the survey found

Primary research surveying 21 project and IT management professionals across construction, IT and manufacturing, as part of the MSc Engineering Management dissertation (2024). Each figure below is the share of respondents who agreed with the statement.

81%

Communication efficiency: agreed AI can significantly improve communication productivity in projects.

67%

Stakeholder understanding: agreed NLP advances will improve the depth of insight into stakeholder demands.

62%

Governance transparency: believe AI enhances stakeholder transparency and the quality of decision data.

52%

Phased adoption: preferred a staged rollout, piloting AI at small scale first to build cultural acceptance.

05 Roadmap

Strategic implementation

Phase 1 · Descriptive automation

Automating low-value tasks: meeting transcriptions, schedule syncs and status report generation.

Phase 2 · Predictive intelligence

Sentiment analysis and machine learning to forecast resource strain and communication lag before they escalate.

Phase 3 · Prescriptive governance

AI suggesting specific engagement strategies for high-priority stakeholders, based on historical trend data.

06 Download

Read the full research abstract

Scroll to Top