The full article and learnings are available to DataIQ clients on our members only hub.
Monotype’s data estate was fragmented by acquisitions, duplicate systems, and shadow data lakes. Its response was not a large MDM overhaul, but a focused, iterative push toward trusted, AI-ready data.
Agentic MDM represents a monumental shift from rule-based MDM to a system where AI agents can autonomously manage, clean, and synchronize data across numerous platforms in real time. Agentic MDM distributes intelligence across systems, rather than a classic centralized manual approach, which means increased self-governance, adaptability, and decision intelligence without continuous human intervention.
The fireside chat, held at the DataIQ 100 Summit in Nashville, explored how Monotype used Syncari’s agentic MDM platform to unify customer data, automate cleansing, and improve operational workflows.
AI adoption is increasing pressure on already-fragmented data environments. Monotype’s experience shows how data leaders can move effectively by proving value through narrow, business-facing use cases before successful scaling.
- How do you modernize MDM without a long, expensive implementation?
- How can acquisition-heavy organizations unify fragmented systems?
- What early use cases prove value quickly?
- How can trusted data improve sales, customer insight, and AI readiness?
- What lessons emerge from iterative, low-code data management?
Prove value before expanding
Carol Vasington Lee, VP, Global Data, joined Monotype as its first dedicated data employee and found no usable master data foundation upon her arrival. Data silos were reality with marketing, product, and other teams each building their own data environments and dashboards.
Rather than launch a broad MDM transformation, Monotype began with a practical acquisition-driven problem focused on bringing together multiple marketing and solution provider platforms. The goal was to create enough value early that the platform justified itself and, as Carol put it, “If this is all we do, we’ll have gotten value for our money.”
Use customer data as the first business case
The first major win was answering a basic question Monotype could not previously answer: how many customers do we have? Jonathan Goldberg, Director, Product Strategy at Syncari, noted that “the most common MDM problem is always ‘who are your customers?’”
Monotype built an account 360 view by blending CRM data and feeding in purchase data from an e-retailer platform, including around 8 million records. This created visibility into overlap between B2B clients and B2C buyers, turning master data from a back-office clean-up exercise into a commercial asset.
AI readiness is a data infrastructure problem
Monotype’s CEO encouraged the leadership team to build with AI. Carol saw both sides of that push as it is great to receive the support, but this now means that new pressures are created, and has been repeatedly shown, executives are eager to see ROI on AI. The issue Carol faced at the start was that teams were building agents against whatever data they could access, creating more silos and more inconsistent insight.
Her response was to focus on “data infrastructure; that trusted, curated data set” that teams could integrate against. The lesson is that AI experimentation without governed, reusable data foundations catalyses fragmentation.
Stage logic before hard-coding
When Monotype revamped its Salesforce account type logic, it used agentic MDM to stage rules and test them against every record before involving the Salesforce team. This exposed edge cases early and avoided sprint-by-sprint rework. Data and AI leaders should look to use MDM and orchestration layers as a safe testing ground before embedding logic in operational platforms.
Optimize for steady progress
Carol’s clearest lesson was “don’t try to boil the ocean at once” when it comes to implementing agentic MDM. Monotype had multiple use cases running, but each was layered onto a practical business need.
A final mindset shift for data and AI professionals was to consider that, in the age of AI and no-code tools, teams need to evaluate what constitutes good enough to move forward. That means creating momentum while improving controls over time.
The full article and learnings are available to DataIQ clients on our members only hub.
