Appendix 3 — Monitoring & Analytics Crosswalk

(From Raw Signals to Institutional Decisions)

Purpose of this appendix
To make explicit how monitoring systems, data streams, and informal signals are translated into decision-relevant intelligence—and to prevent three recurring failures:

1. confusing activity tracking with situational awareness,
2. mistaking volume for significance, and
3. outsourcing judgment to dashboards.

This appendix answers a practical question SPES faces daily:

“We are seeing this — so what does it actually mean for leadership?”

3.1 Core Principle: Monitoring Is Not Insight

SPES monitoring is:

• interpretive, not extractive
• selective, not exhaustive
• decision-oriented, not performance-oriented

Any monitoring stream that does not inform a real decision is noise and should be deprioritized.

3.2 Monitoring Domains & System Sources

SPES monitoring spans six domains, each with different tools, reliability, and risks.

3.2.1 Traditional Media Monitoring

Sources

• Local print press
• TV and radio transcripts
• Online news portals
• News aggregators (manual or automated)

What is monitored

• Headline framing
• Quotation of Embassy language
• Repetition across outlets
• Shift from reporting to commentary

What SPES extracts

• framing trends (not sentiment scores)
• attribution patterns
• narrative escalation signals

Common failure

Treating media volume as hostility rather than curiosity or amplification.

3.2.2 Social Media Monitoring (Embassy-Owned)

Platforms

• Facebook
• X (Twitter)
• Instagram
• YouTube

Raw signals

• post frequency
• engagement counts
• comment volume
• share velocity

SPES interpretation focuses on

• what kind of engagement (question, grievance, mobilization)
• time-to-escalation
• policy attribution language in comments

SPES explicitly avoids

• follower obsession
• sentiment automation
• demographic inference beyond aggregates

3.2.3 Social Media Monitoring (External)

Includes

• influential local accounts
• journalists’ feeds
• activist amplification nodes
• rumor-spreading clusters

SPES tracks

• issue appearance timing
• phrasing reuse
• misattribution of Embassy positions

SPES does not

• map networks
• track individuals
• infer coordination without evidence

3.2.4 Programmatic Monitoring (Internal Signals)

Sources

• Program officers
• American Spaces staff
• LES informal reporting
• Partner feedback

What SPES listens for

• discomfort
• confusion
• sudden caution
• “people are asking about…”

These signals often precede public escalation.

3.2.5 Exchange & Alumni Visibility Monitoring

Sources

• media mentions
• social posts
• event participation
• public commentary

What SPES tracks

• visibility spikes
• attribution language (“U.S.-trained”)
• politicization cues

What SPES does not do

• evaluate alumni behavior
• advise distancing absent risk

3.2.6 Washington & Policy Signal Monitoring

Sources

• cables
• press briefings
• policy speeches
• travel advisories

SPES role

• flag misalignment between Washington posture and local messaging
• anticipate sensitivity shifts

3.3 From Monitoring to Analysis: The Crosswalk

This section makes the conversion explicit.

3.3.1 Signal → Question → Decision Chain

Example

Signal observed

• Embassy Facebook post shared widely with hostile comments.

SPES question

• Is this confusion, grievance, or mobilization?

Decision affected

• Whether to clarify, pause, or disengage.

3.3.2 Crosswalk Table — System to Decision

3.4 Metrics SPES Uses (and Why)

SPES uses only metrics that answer a question.

Commonly used

• posting rate
• engagement medians (not maxima)
• time-to-escalation
• pickup frequency
• repetition of phrases

Rarely used

• follower growth
• sentiment percentages
• reach alone

Never used

• individual-level demographics
• predictive scoring
• “influence scores”

3.5 Sampling Logic (Defensibility Without Exhaustion)

SPES rarely reviews everything.

Acceptable sampling:

• top engagement posts per quarter
• posts tied to sensitive topics
• defined time windows around incidents

Sampling must be:

• documented
• repeatable
• explained

3.6 Output Types Generated From Monitoring

Monitoring feeds into:

• interim leadership briefs
• risk register updates
• holding language drafts
• escalation recommendations
• post-mortem notes

It does not feed:

• performance evaluations
• disciplinary action
• audience targeting

3.7 Failure Modes This Appendix Prevents

Without this crosswalk:

• dashboards drive decisions
• anecdotes dominate
• staff argue about “what the data says”
• monitoring becomes surveillance

With it:

• monitoring supports judgment
• disagreement is structured
• leadership sees implications, not charts