RedCloud CEO Declares the Decline of Systems of Record as It Validates R.A.I.D. AI Engine Built for Modern Global Trade

Rhea-AI Summary

RedCloud (NASDAQ: RCT) validated its R.A.I.D. (Realtime AI for Distribution) engine on 3.7 million completed FMCG transactions, reporting ranking accuracy (NDCG) significantly above the commercial-impact threshold of 80%. The company plans to commercialize R.A.I.D. within the RedAI platform later in 2026, embedding AI at the point of transaction to drive decisions across forecasting, replenishment, pricing, credit and route optimisation.

RedCloud cites an eight-year proprietary transaction dataset from emerging markets as the foundation for R.A.I.D.'s performance and says the system includes confidence scoring and fallback mechanisms for lower-data environments.

Positive

• 3.7 million completed FMCG transactions used to train R.A.I.D.
• R.A.I.D. achieved ranking accuracy (NDCG) significantly above 80%
• Commercial rollout planned within RedAI later in 2026
• Eight-year proprietary transaction dataset across emerging markets

Negative

• None.

News Market Reaction – RCT

+9.32%

5 alerts

+9.32% Session close to close

+4.9% Peak Tracked

-8.5% Trough Tracked

$42.36M Market Cap

0.9x Rel. Volume

In the Mar 23 session, RCT gained 9.32%, reflecting a notable positive market reaction. Argus tracked a peak move of +4.9% during that session. Argus tracked a trough of -8.5% from its starting point during tracking. Our momentum scanner triggered 5 alerts that day, indicating moderate trading interest and price volatility.

Data tracked by StockTitan Argus on the day of publication.

Market Context

The stock moved +9.3% in the session following this news. A strong positive reaction aligns with Red...

Analysis

The stock moved +9.3% in the session following this news. A strong positive reaction aligns with RedCloud’s strategy of turning its proprietary FMCG transaction data into agentic AI products. Past AI-tagged news produced mixed but sometimes sharp upside moves, such as double-digit gains after key RedAI announcements. Investors would still need to weigh financing structures disclosed in recent filings and the stock’s position far below its 5.36 52-week high when assessing durability.

Key Figures

Live FMCG transactions: 3.7 million transactions Accuracy threshold: >80% NDCG Order size example: 500 cases +1 more

4 metrics

Live FMCG transactions 3.7 million transactions Data used to validate R.A.I.D. accuracy

Accuracy threshold >80% NDCG Threshold cited as commercially impactful for ranking systems

Order size example 500 cases Illustrative distributor order showing executed intent in training data

Data history 8 years Duration over which proprietary transaction dataset was accumulated

Previous AI Reports

5 past events · Latest: Mar 03 (Positive)

Same Type Pattern 5 events

Date	Event	Sentiment	24h Move	Catalyst
Mar 03	AI network milestone	Positive	-8.0%	Surpassed 100,000 customers and activated agentic AI layer across network.
Jan 12	AI product launch	Positive	+12.1%	Early access launch of RedAI Trading Co-Pilot for live FMCG trading.
Dec 11	AI-linked earnings	Positive	+8.3%	H1 2025 results, $100M 2026 target and AI-driven infrastructure expansion.
Dec 02	AI platform expansion	Positive	-9.0%	Upgraded Red101 retail trading experience powered by RedAI targeting inventory gap.
Nov 25	AI thought leadership	Positive	-1.7%	CEO participation in global AI summits and reiteration of agentic AI focus.

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Pattern Detected

AI/agentic AI announcements have often been framed positively but produced mixed reactions, with more divergences than alignments between upbeat AI news and next-day price moves.

Recent Company History

Over the past year, RedCloud has repeatedly highlighted its AI and agentic infrastructure strategy. Prior AI-tagged events featured customer milestones, RedAI product launches, large FMCG market targets and deleveraging, alongside executive visibility at major AI conferences. Price reactions to these AI updates were inconsistent, with both double-digit gains and steep declines. Today’s validation of the R.A.I.D. engine and its accuracy on 3.7 million transactions fits this ongoing push to commercialize AI across global trade networks.

Key Terms

erp systems, systems of record, agentic infrastructure, learning-to-rank, +3 more

7 terms

erp systems technical

"“ERP systems were masterpieces of documentation,” said Justin Floyd"

ERP systems are integrated software platforms that help organizations manage and coordinate their core business activities, such as finance, supply chain, human resources, and operations, in one unified system. They enable smooth information flow across departments, improving efficiency and decision-making. For investors, ERP systems indicate how well a company manages its resources and operations, which can impact its overall performance and profitability.

systems of record technical

"From Systems of Record to Systems of Action"

Systems of record are the primary IT applications or databases that store an organization’s official, auditable copies of critical information—such as financial transactions, ownership records, or customer details. Investors care because these systems are the dependable source used for reporting, compliance, and audits; like a company’s official ledger or filing cabinet, their accuracy affects reported results, legal risk, and the reliability of management’s decisions.

agentic infrastructure technical

"move to agentic infrastructure - systems that observe continuously"

Agentic infrastructure is the collection of software, hardware, networks and rules that let autonomous digital agents—software programs that can make decisions and act—operate reliably at scale. It matters to investors because it is the foundation that enables automation, new revenue streams and efficiency gains, while also concentrating technological and regulatory risks much like roads and traffic signals determine how cars can move and where congestion or accidents may occur.

learning-to-rank technical

"R.A.I.D. is a two-tier, learning-to-rank system"

A method of machine learning that teaches a computer to put a list of items in the best order for a given purpose, such as search results, recommendations, or risk scores. Like a librarian who learns from readers’ choices which books to display at eye level, it helps products surface the most relevant items quickly; for investors, that can boost user engagement, ad or sales revenue, and the quality of automated decisions while also creating operational and regulatory risks to monitor.

confidence scoring technical

"The system also incorporates confidence scoring and fallback mechanisms"

A numerical assessment that expresses how certain a model, test, or decision is about a particular result or prediction. Think of it like a weather app saying there’s an 80% chance of rain—higher scores mean more trust in the outcome, while lower scores signal more uncertainty or need for caution. For investors, confidence scoring helps gauge the reliability of forecasts, diagnostics, or automated decisions and guides how much weight to give that information in portfolio, risk, or deal-making choices.

fallback mechanisms technical

"incorporates confidence scoring and fallback mechanisms, enabling performance to degrade"

A fallback mechanism is a planned backup or alternative process that a company uses when its main system, product, supply route, or approval path fails or becomes unavailable. Like a spare tire or alternate route on a trip, it helps keep operations running, protect revenue, and limit losses; investors look for clear, tested fallbacks because they reduce the chance of sudden disruption and show management prepares for foreseeable risks.

agentic ai technical

"The Agentic Era Begins As enterprises continue to invest in AI layered onto legacy systems"

Agentic AI refers to computer systems that can make their own decisions and take actions without needing someone to tell them what to do each time. It's like giving a robot a degree of independence to solve problems or achieve goals on its own, which matters because it could change how we work and interact with technology in everyday life.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

R.A.I.D. (Realtime AI for Distribution) outperforms industry benchmarks for accuracy on 3.7 million live FMCG transactions, redefining decision making in volatile global markets

London, UK, March 23, 2026 (GLOBE NEWSWIRE) -- RedCloud Holdings plc (NASDAQ: RCT) ("RedCloud" or the "Company"), the technology company building AI infrastructure for global trade today declared that the era of traditional ERP systems faces a steep decline in relevance, alongside validating R.A.I.D. (Realtime AI for Distribution), its AI engine embedded within the flow of trade, delivering accuracy levels that outperform industry benchmarks1 across 3.7 million live FMCG transactions in market conditions defined by volatility, disruption and constant change. 

From escalating tensions in the Middle East disrupting critical trade corridors, to tightening credit conditions and rapidly shifting demand patterns, the Company believes the operating environment for global trade has fundamentally changed. Supply chains are no longer linear. Signals are no longer stable. Decisions can no longer wait. Yet the systems used to manage global trade were not built for this reality. 

“ERP systems were masterpieces of documentation,” said Justin Floyd, CEO of RedCloud Holdings plc. “They were built to record transactions, enforce process and explain what already happened. But global trade no longer pauses between decisions. It runs on volatility, probability and overlapping signals. Software designed to memorialise the past cannot govern markets that move continuously.” 

From Systems of Record to Systems of Action 

For decades, enterprise software has operated as a system of record: humans initiate action, software responds, and decisions move sequentially. That model worked when information moved slowly and supply chains were predictable. It does not work in an economy defined by fragmentation, uncertainty and constant motion. 

Demand shifts daily. Credit conditions change mid-cycle. Distributors must act without waiting for perfect information. In this environment, latency is not inefficiency, it is likely lost margin. 

“The problem isn’t AI,” Floyd continues. “The problem is what enterprise software was built to do. You cannot bolt intelligence onto architectures designed to ask permission at every step. Intelligence must live inside the infrastructure itself.” 

RedCloud defines this shift as the move to agentic infrastructure - systems that observe continuously, act within defined guardrails and optimise decisions as conditions change. 

Proof, Not Theory 

R.A.I.D. is a two-tier, learning-to-rank system, designed to deliver highly relevant product recommendations at scale across distributor networks. Trained on 3.7 million completed trade transactions within RedCloud’s RedAI platform, R.A.I.D. has achieved accuracy levels (NDCG) that are significantly above the >80% threshold generally considered commercially impactful for ranking systems. 

Unlike consumer recommendation engines that infer intent from behavioural signals, R.A.I.D. is trained on executed trade.  These are real orders, real volumes, real commercial outcomes. 

“This is not inferred behaviour. This is real-world data,” said Floyd. “A distributor ordering 500 cases of a product is not a signal. It is intent executed. That changes what AI can do. Global FMCG trade generates millions of commercial decisions every day. The intelligence layer required to support those decisions at scale has not previously existed.” 

R.A.I.D. is not designed to generate reports. It is designed to drive decisions. By embedding intelligence directly at the point of transaction, RedCloud enables distributors and retailers to act on predicted outcomes, increasing average order value, improving repeat rates and optimising inventory flow in near-live trading environments. This represents a shift from backward-looking analysis to forward-looking execution. 

One Engine. Multiple Decisions. Continuous Impact. 

R.A.I.D. is designed to operate across multiple commercial decision layers simultaneously, including: 

• Demand forecasting 

• Replenishment optimisation 

• Stockout prevention 

• Pricing and promotional effectiveness 

• Credit and risk signals 

• Route and distribution optimisation 

The R.A.I.D. engine will be commercialized within the RedAI platform later this year and exist within the intelligence layer embedded in RedAI trading networks. The system also incorporates confidence scoring and fallback mechanisms, enabling performance to degrade gracefully in lower-data environments, supporting expansion across new markets and datasets. 

The architecture is designed to scale across geographies and operate on both RedAI’s aggregated dataset and partner-specific data environments. 

RedCloud's proprietary transaction dataset — accumulated over eight years across emerging markets where conventional enterprise software has been unable to operate at scale — represents the data foundation underlying R.A.I.D.'s performance. As disruption to global trade corridors intensifies, demand for real-time distribution intelligence, that dataset, and the accuracy it enables, is difficult to replicate. 

The Agentic Era Begins 

As enterprises continue to invest in AI layered onto legacy systems, many are struggling to translate experimentation into measurable commercial outcomes. RedCloud believes the issue is structural. “The era of software you use is ending,” Floyd concluded. “The era of infrastructure that acts has already begun.”

About RedCloud Holdings plc 

RedCloud’s mission is to build the intelligence infrastructure of global trade, through generation and aggregation of proprietary trading and market data from across the FMCG industry through its RedAI platform (“RedAI”). RedCloud provides market intelligence based on proprietary trading data across categories in each of its markets. The Company also delivers a ‘trading layer’ of products for use by its customers, to enable intelligent digital exchange of everyday consumer supplies of FMCG products across business supply chains, supported by a payments and lending ecosystem intended to streamline trade. RedCloud believes its platform and associated products and services solve a decades-old problem of how to digitize trade at scale to generate data and intelligence to enable brands, distributors and retailers to maximize business performance across categories in high growth consumer markets.  

RedCloud is a British company registered in London, co-founded by British serial entrepreneur Justin Floyd and Algerian born Soumaya Hamzaoui. In March 2025, RedCloud achieved its initial public offering on New York’s Nasdaq stock exchange under the ticker “RCT”. For more information about RedCloud and its platform, please visit www.redcloudtechnology.com and connect on LinkedIn.

Footnotes 

1 Normalized Discounted Cumulative Gain (NDCG) is a popular metric used to evaluate the quality of ranking in search engines and recommendation systems. Unlike simpler metrics that only check if a result is "relevant or not," NDCG accounts for both the relevance of items and their specific positions in a list. 

Forward-Looking Statements 

The information in this press release may include forward-looking statements within the meaning of the federal securities laws. These statements generally relate to future events or our future financial or operating performance, including, but not limited to, statements regarding future financial performance, growth, business strategies, market opportunities, and operational outlook, RedCloud’s ability to successfully integrate the R.A.I.D. engine into its platform, the continued accuracy of the R.A.I.D. engine’s recommendations to users of RedCloud’s platform and expectations regarding the R.A.I.D. engine’s ability to maintain performance levels exceeding the generally recognized commercial impact threshold of 80% accuracy for ranking systems. When used in this press release, words such as “expect,” “project,” “estimate,” “believe,” “anticipate,” “intend,” “plan,” “seek,” “forecast,” “target,” “predict,” “may,” “should,” “would,” “could,” and “will,” the negative of these terms and similar expressions are intended to identify forward-looking statements, although not all forward-looking statements contain such identifying words. Forward-looking statements are based on management’s current expectations and assumptions, and are subject to inherent uncertainties, risks and changes in circumstances that are difficult to predict, including, but not limited to, the completion of the Company’s year-end financial closing procedures, the preparation and audit of the Company’s financial statements, potential adjustments to preliminary results, changes in economic, market, or industry conditions, customer demand, competitive pressures, regulatory developments, the successful execution of its joint venture strategy and launches, the ability to expand across operational markets, the ability to leverage RedCloud’s future global developments in the use of AI and associated supply chain models through RedAI, and whether RedCloud’s upcoming product release meets customer expectations leading to broad adoption. As a result, actual results could differ materially from those indicated in these forward-looking statements. When considering these forward-looking statements, you should keep in mind the risk factors and other cautionary statements in RedCloud’s described in “Cautionary Note Regarding Forward-Looking Statements,” “Item 3. Key Information – D. Risk Factors” and “Item 5. Operating and Financial Review and Prospects” in RedCloud’s Annual Report on Form 20-F for the year ended December 31, 2024, which was filed with the Securities and Exchange Commission (the “SEC”) on May 16, 2025, as well as other periodic reports and documents filed by the Company with the SEC. The preliminary financial information included in this press release is unaudited, subject to change, and may differ from final results to be reflected in the Company’s audited financial statements. RedCloud undertakes no obligation and does not intend to update these forward-looking statements to reflect events or circumstances occurring after this press release. You are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date of this press release. Information contained on, or that can be accessed through, the Company's website or any other website or any social media is expressly not incorporated by reference into and is not a part of this press release. 

Contacts: 

Investor Relations: Investor.relations@redcloudtechnology.com  

Media Relations: media@redcloudtechnology.com 

FAQ

What accuracy did RedCloud (RCT) report for R.A.I.D. on 3.7 million transactions?

R.A.I.D. achieved ranking accuracy (NDCG) reported as significantly above the 80% commercial-impact threshold. According to the company, this performance was measured on 3.7 million completed FMCG trade transactions inside RedAI.

When will RedCloud (RCT) commercialize the R.A.I.D. engine within RedAI?

RedCloud plans to commercialize R.A.I.D. within the RedAI platform later in 2026. According to the company, the engine will sit in the intelligence layer and be embedded at the point of transaction for distributors and retailers.

What decision areas does R.A.I.D. target for RedCloud (RCT) customers?

R.A.I.D. targets multiple commercial decision layers including forecasting, replenishment, pricing, credit signals, stockout prevention and route optimisation. According to the company, the engine operates across these layers simultaneously to drive near-live trading decisions.

How did RedCloud (RCT) train R.A.I.D. differently from consumer recommenders?

R.A.I.D. was trained on executed trade — real orders and volumes — rather than inferred behavioural signals common to consumer recommenders. According to the company, this uses completed commercial outcomes to reflect true buying intent at scale.

What dataset underpins R.A.I.D.'s performance for RedCloud (RCT)?

R.A.I.D.'s performance is built on RedCloud's proprietary transaction dataset accumulated over eight years across emerging markets. According to the company, this dataset represents the foundation for training and scaling the engine.

How does R.A.I.D. handle lower-data or new-market environments for RedCloud (RCT)?

R.A.I.D. incorporates confidence scoring and fallback mechanisms to degrade performance gracefully in lower-data settings. According to the company, this design supports expansion across new geographies and partner-specific datasets.