From the headline to your bottom line, in six steps or fewer.
Cross-domain cascade analysis, with every claim cited. Elite foresight for teams without a retained advisor.
An EU power-market rule today could reshape what it costs to run an AI data centre.
Example chains. Illustrative of the platform's reasoning, not surfaced predictions. Live chains have shipped weekly since W19; the public verdict ledger is live, with the first horizons now closing.
Six steps. Or fewer.
In 1973, the sociologist Mark Granovetter argued that the most valuable information moves through weak ties between clusters, not the dense ties inside them. Fifty years later, it still holds. What's new: an AI system can now trace those ties at scale, on demand, without a research department.
Any global event reaches your local exposure in six steps or fewer. We find those steps.
Cross-domain impact analysis is what elite consultancies sell for six- and seven-figure retainers. Maps of how a regulatory shift in Brussels lands in industrial input costs in Eindhoven, then in consumer brand M&A in New York. Chains of cause and effect, built from public information, then sold back to clients who could build them in-house if they had the time, the team, and the patience.
The information is largely public. The deals are not. We surface the chains, with every claim cited. Robin Hood, for strategic foresight.
"The strength of weak ties," Granovetter, AJS, 1973
Profile. Diagnostic. Cascades.
Chainge turns your structured exposure profile into a weekly cross-domain cascade briefing. The profile runs in your own AI tool, a diagnostic challenges it, and from then on the global signal corpus runs against your specific shape every week. Behind the briefing: a four-stage AI pipeline wired to primary sources (EUR-Lex, FRED, GDELT, Polymarket). It generates candidate cascades across your full exposure surface, then gates them hard, leading with a few high-conviction headlines and marking the deeper field beneath. Every claim cited. Every chain scored.
For the people without an advisory team.
McKinsey serves the Fortune 500. We serve everyone else. Operators building strategy on the back of a napkin. Mid-market teams who can't justify a six-figure retainer. Analysts who want a methodology layer they can show their clients.
The verdict ledger is live.
Every prediction in a chain runs to a verdict horizon, capped at 90 days. When the horizon arrives, we mark it: right, partial, wrong, or unclear. Mechanical, not narrated. No retroactive editing.
live weekly, W19 through W29. first prediction window closed 2026-06-01; first node checks resolved 2026-06-11; terminal horizons rolling forward as they close.
W25 · 2026-06-15 NL grid flips from net-export to net-import: public Dutch capacity data already shows offtake congestion exceeding feed-in, so reinforcement is committed while the timing of the regime shift stays the open question. →
W24 · 2026-06-08 A REACH SVHC candidate listing would pressure EU OEM revenue: the one chain in the set that resolves mechanically against the public ECHA register, no operator judgment. →
W23 · 2026-06-01 EU electricity catch-22: the same EED efficiency-first principle ACM uses to push non-wire alternatives can be turned to disallow a DSO's reinforcement-capex hedge. →
One profile. One diagnostic. One chain.
Anne runs Brabant Advanced Materials, a specialty chemicals SME in Eindhoven. We took her through profile ingest, a diagnostic, and the cascade that came out, including the regulation the diagnostic caught that she hadn't listed and the German lobbying fight that decides her outcome.
Why this can't be copied.
The AI part is replicable. Anyone with an API key can build a version of this. The moat sits behind it: primary data sources, a public methodology track record, and a contribution flywheel that compounds with use.
-
01 · data flywheel
Contribution-based pricing. Users who feed signals back pay less. The platform compounds with use. Wikipedia meets prediction markets, for strategic intelligence.
-
02 · methodology track record
Every chain timestamped, every prediction publicly verifiable. The track record compounds. The longer we run, the harder we are to displace.
-
03 · primary data, not headlines
Regulatory filings, central bank releases, shipping data, commodity prices, satellite imagery. Reasoning from structured primary sources is the credibility gap. It separates Chainge from any wrapper around a chatbot.
Built by Niels Langereis.
Eighteen years building platforms, integrating businesses, and shipping technology, across PE-backed groups and global agency networks. The longer Niels did it, the harder one observation got to ignore: most of what consultancies sell as proprietary analysis isn't. It's labor, on public information, priced for institutional buyers. Chainge automates the labor and prices it for everyone else.
Most recently, group-wide AI governance and platform strategy at Normec Group across 70+ entities. Before that: CEO at New Story, CEO at Hogarth (WPP) Netherlands, Managing Director at Wavemaker (WPP), and earlier senior roles at Publicis (Amsterdam, Dubai).
In 2004, the same year James Surowiecki popularized the idea in The Wisdom of Crowds, Scott Page and Lu Hong published a result that Page himself didn't initially believe: a randomly chosen, diverse group of problem-solvers consistently outperforms a homogeneous group of the best individual experts. He ran the simulations again before publishing.
A younger Niels had heard a version of this years before, told as a story about a research team in European banking that supposedly mixed economists and lawyers with historians, anthropologists, and political experts, all looking at the same question from a different side of the table. He was never able to verify it, but he also never stopped thinking about it. His master's thesis on creative mass collaboration argued the same point in a different domain. Chainge is that room, opened up.
You start with a profile and a direct conversation. We run the diagnostic and the weekly cascade against your specific exposure from there.
See how the world reaches your exposure.
Start with a profile of your company and a direct conversation. We run the weekly cascade against your specific exposure, with every claim cited.
talk to niels →We use your email only to send Chainge updates. No third parties. Unsubscribe anytime. Privacy notice.