Platform · METI

When should it move?

A predictive forecasting engine that fuses time-series analysis with NLP-driven sentiment tracking from news and policy statements to identify optimal entry windows.

1–6: month forecast windows
68: day window identified in Rwanda, Q2 2026
3: postures: enter, hold, watch

The Kilwa METI view: Kenya scored 74 and marked OPTIMAL, with a score trend chart, a six-factor market comparison grid, key drivers and a dated catalyst queue. — METI in the platform. Illustrative output — scores and figures shown are sample values, not live data.

Predictive timing intelligence

What METI is

The Market Entry Timing Index (METI) is Kilwa's timing model for frontier markets. It fuses time-series analysis with multilingual news and policy sentiment to forecast 1 to 6 month entry windows, returning an ENTER, HOLD or WATCH posture with a confidence indicator for each market covered.

Suitability without timing is half an answer. METI forecasts 1–6 month entry windows by combining structured time-series signals with multilingual sentiment drawn from local news and policy sources.

The output is an explicit posture — ENTER, HOLD or WATCH — attached to a window and a confidence indicator, rather than a directional opinion the reader has to translate into a decision.

METI is the reason our East Africa brief could put a 68-day window on Rwandan digital infrastructure rather than describing the market as 'attractive'. The posture is a property of the market under a published methodology, not a recommendation to any reader.

Capabilities

Hybrid forecast
The Sentiment Pulse delta enters a time-series forecast as an exogenous variable, so a high-frequency signal informs a low-frequency one instead of competing with it.
Learned weighting
A Bayesian optimisation process tunes how hard the sentiment signal is weighted, rather than fixing it by assumption.
1–6 month windows
Forecast horizons matched to how institutional deployment decisions are actually scheduled.
ENTER / HOLD / WATCH
An explicit posture per market, with a confidence indicator attached.
Sentiment fusion
Structured time-series and unstructured multilingual sentiment enter the same model.
Evaluated against outcomes
Windows are evaluated against realised outcomes, and the evaluation method is published with the model card.

Seen in the research

The Four Percent Trap

A Kilwa reading of the World Bank's April 2026 outlook, extended to all 54 markets, scored for distance from takeoff, and repriced for the war premium.

Read the analysis

2.0%per-capita growth at the recovery's peak. The economies that escaped poverty last century compounded at three times that pace