Clay Lab is our internal R&D function. We don't wait for a client project to explore emerging technologies — we build, test, and validate so that when you need it, we've already made the mistakes.
The mandate
Clay Lab exists because client engagements are not the right venue for first experiments. When we bring a technology recommendation to a client, it needs to be backed by real implementation experience — not a vendor brochure or a conference talk.
Our engineers and analysts spend structured time each week running experiments: building proofs of concept, stress-testing architectural assumptions, and validating claims before we make them.
6
Active experiments currently running
18+
Proofs of concept completed since 2020
3
Lab outputs promoted to client engagements in 2024
Current experiments
Training a lightweight classification model on 3 years of water billing data to flag anomalous consumption patterns before they become disputes. Current accuracy: 91.3% on held-out validation set.
Building a prompt-to-insight interface that lets non-technical users ask questions in plain English and receive structured Qlik Sense dashboard responses. Currently in internal beta.
Exploring whether ARIS process mining exports can be parsed and converted into ServiceNow workflow scaffolding automatically — reducing implementation time on App Engine builds.
Developed an internal framework for automatically scoring the data quality output of Talend ETL pipelines across three dimensions: completeness, consistency, and timeliness. Now used on all client engagements.
Designing a reference architecture for state-level government departments that need enterprise-grade data infrastructure on constrained budgets.
Built a library of reusable Qlik Sense extensions for visualising infrastructure assets on maps — water pipelines, road maintenance, utility networks.
Focus areas
Practical enterprise applications beyond the hype cycle — from document processing to natural language interfaces on existing data platforms.
Using ARIS and Celonis-style tooling to understand how organisations actually work before recommending how they should change.
Evaluating AWS, Azure, and GCP data services for enterprise and government workloads where cost, governance, and resilience all matter.
The discipline of deploying, monitoring, and maintaining predictive models in production — especially in regulated environments.
Microservices, API-first design, and modular system architecture for government ERP and enterprise platforms that need to evolve over time.
Early exploration of multi-step AI agents that can operate within enterprise systems — ServiceNow, Qlik, and custom platforms.
Collaborate with Clay Lab
We're occasionally open to early-stage R&D partnerships where clients want to explore a new technology approach before committing to a full engagement.
