A Digital Future for Insurance: Embedding Data at the Core of Commercial Lines Underwriting

Breaking Through the Challenges of Commercial Underwriting

Underwriting is the heart of insurance. The ability to effectively analyze and price risks distinguishes good underwriters from great ones, and profitable insurers from unprofitable ones. Historically, underwriting has been part art, part science—based on technical models, exposure information, and risk appetite, but with heavy subjective interpretation. Until now, this approach was justified by shortfalls in available data and computing power. Today, however, there is an abundance of both, offering the opportunity to be much more scientific—particularly in commercial lines, where risks are the most complex and diversified.

Yet, commercial lines underwriting faces many barriers. Current processes are often fragmented, inefficient, reliant on poor-quality data, and slow to change. For example:

Fractured, messy submissions require rekeying and categorization. Multiple studies indicate that insurers spend more than 40 percent of their time performing non-underwriting activities, such as rekeying data, generating documents, and supporting sales or marketing.
Lack of real-time Management Information (MI) for efficient prioritization limits visibility into concentration and trends across sector, geography, and client. This leads underwriters to prioritize simpler or “louder” requests, rather than those where the business wants to win margin or expand their footprint.
Fragmented decisioning processes involve multiple handoffs across CRMs, Excel-based rating models, emails, forms, third-party databases, and internal systems.
Low-quality and siloed data is cumbersome to join together, and by the time it is, the market may have already moved on.
Slow speed of change to add new modeling factors or data sources, often driven by the complexity of large, slow-moving monolithic systems.

The Opportunity: Embedding Data at the Core of Underwriting

While incumbent commercial insurers tackle their process and data challenges, new startups are evolving and taking share. Forward-thinking insurers are investing in new underwriting solutions and data platforms, improving their use of data and risk selection, and finding new models for underwriting.

The nature of risks is also changing. While economic concerns dominated 10 years ago, technology, extreme weather, and climate change prevail today. Carbon transition, AI products, and increased natural catastrophe frequency all require new product innovation, new data sources, and an efficient pace to seize the space.

In the face of legacy underwriting processes, the changing nature of risk, and an evolving competitive landscape, commercial incumbents face a clear impetus to transform the underwriting process and embed data at its core.

By embedding data at its core, we mean viewing data as a strategic lever of growth (rather than just a set of tools) and developing a clear view of the strategy and capabilities needed to support it.

Better use of data can offer improved value, speed, and quality:

Value: Improved risk selection, response times, and conversion rates; greater underwriter productivity; increased risk mitigation at reduced frictional cost to society.
Speed: Improved cycle times and faster product innovation.
Quality: Higher decision-making quality (translating into improvements in loss ratio) and reduced risk of errors at both the risk and portfolio level, with better broker experience.

The Pathway to Modernizing Underwriting

There are three key steps to placing data at the core of underwriting:

Strategy: Identifying the points of leverage for data to accelerate and improve underwriting.
Capabilities: Building the infrastructure, analytics, operations, and people needed to support this.
Pathway: Defining the necessary transition states to maturity and embedding continuous improvement, rather than a one-and-done delivery.

Strategy: Points of Leverage

To start, business, technology, and data teams need to come together to identify the points of leverage that data can offer across the underwriting process. Key questions to ask include:

How can our underwriting processes be improved by leveraging data?
How can we improve the quality and consistency of data collected?
What kinds of decisions do we make today, and what data can we bring to make them better informed?
How would we change our decision-making if we had 1,000 times the data and the insights from them readily available at our fingertips?
Where can we make our business operations more efficient through automation?

There are five main ways the underwriting journey can leverage a data-centric platform:

Automated ingestion and triaging of submissions
Enhanced underwriting and decisioning
Real-time portfolio steering and scenario management
Continuous improvement of decisioning and data ingestion
New product innovation

Automated Ingestion and Triaging of Submissions

Broker submissions across channels are automatically read and prioritized based on the likelihood of conversion and renewal, and comparison against concentration limits.
Open cases are intelligently triaged to the most appropriate underwriter based on historical precedent with the risk type and broker, as well as current portfolio status. Cases with a low likelihood of quoting or conversion are deprioritized.

Enhanced Underwriting and Decisioning

The overall process is governed by a workflow that minimizes manual handoffs between people and systems.
Underwriters are presented with priced, decision-ready risks that have been ingested, categorized, and passed through decision intelligence engines.
AI-generated contextual prompts give underwriters relevant advice from manuals and show how peers have handled the decisioning and pricing of similar risks.
Underwriters receive requests to update coverage or add endorsements in the same way as a decision-ready risk, coupled with an AI-generated recommendation and a view of the expected portfolio impact.

Real-Time Portfolio Steering and Scenario Management

Management has a portfolio management tool that displays a real-time view of exposures and performance against risk appetite, enabling them to prioritize footprint, steer decision-making, and adapt pricing in real-time.
Scenario planning tools allow them to test changes in risk appetite and pricing and preemptively make rate, footprint, and covenant changes.

Continuous Improvement of Decisioning and Data Ingestion

Newly notified or incurred claims feed rapidly into loss figures, with the AI engine generating suggestions to update footprint or pricing in cases of sustained deviations from projected loss rates.
New data sources that can enhance risk understanding are easily added and ingested via APIs to enable continued evolution of pricing and risk appetite.

New Product Innovation

Claims and market data are used to develop and embed new integrated risk prevention and management solutions, built and deployed quickly to test and learn uptake in the market (e.g., parametric solutions, IoT, climate, agricultural risk, etc.).
These benefits will differ for each insurer and team depending on their overall strategy. Identifying the unique set that is relevant will help align business and technology teams around a common vision and outcomes.

Capabilities: Infrastructure, Analytics, Operations, and People

Driven by this compelling opportunity, insurers should evaluate the maturity of their current underwriting and data capabilities. Here are some guiding questions to ask across four key dimensions:

Business Operations: Processes across underwriting and other functions (e.g., finance, risk)

Where are underwriters spending the most time on non-underwriting activities?
Where are handoffs between underwriters, brokers, and other third-party teams?

Infrastructure and Data: Tools and architecture to manage underwriting and use of data

How readily available is submission, pricing, and claims data across teams today?
How differentiated and proprietary is the company’s data?

Analytics: Dashboards and analysis used to generate insights across the company

How accessible is important data by people and systems?
How reliable is that data?
How aligned are teams across the definitions of core metrics (e.g., GWP, retention)?

Team: Maturity of understanding and usage of data across the organization

What is the framework for making operational and strategic decisions today? What data is used, if at all?
How comfortable are team members with using self-serve data tools?

Pathway: The Transition States to Maturity

Set strategy to reorientate around data
Develop focused MMP (/MLP) on new platform
Implement for narrow use case and gain feedback
Scale across other use cases

1. Set Strategy to Reorientate Around Data

The strategic areas of leverage and capability assessment above will inform the priorities for change. Commercial insurers are often most plagued by their infrastructure, which impacts the maturity of data usage across the organization and the ability to change.

There are many Insurtech players and platforms that service different parts of the insurance value chain, but assembling and wiring them up—with data at the core—is key. As seen in banking, there is a trade-off between traditional systems with rich functionality and a long track record (but on aging, inefficient architectures) and new entrants built from the ground up with cloud-based modular architectures and agile delivery that are maturing their breadth and track record.

2. Develop Focused MMP (/MLP) on New Platform

We recommend a “steel-thread” approach to transforming underwriting—picking the smallest end-to-end slice of functionality and building it out thread-by-thread—rather than a monolithic big bang delivery. This allows for rapid end-to-end delivery that proves value quickly and is joined up from end users through to business operations, infrastructure, and data. Early business value can be driven through one of the benefit areas above, accelerating appetite for wider change.

To do this, insurers need to set up a multidisciplinary squad spanning business, technology, and design, empowered to build and make quick decisions. The squad should pick one opportunity, rapidly design the end-to-end “service blueprint,” and bring it to life. For example, building a prototype to automatically ingest documents for a single business line and integrating it into existing workflows.

3. Implement for Narrow Use Case and Gain Feedback

Building the first prototype will require additional effort across the value chain—onboarding new vendors, building cloud, data, and engineering foundations, and establishing new ways of working. Once the first prototype is built, subsequent slices can be onboarded more nimbly and matured over time.

Prototypes often fail to translate into actual results or become embedded into business operations. Realizing value requires establishing clear decision-making frameworks and accountability over key value objectives and results (e.g., number of hours to a decision, percentage of retention). In some cases, this may require organizational realignment around value streams or journeys. Robust data governance frameworks and investment in data quality automation also become key at this stage to enable scale.

4. Scale Across Other Use Cases

Once embedded, the focus should be to industrialize and scale to other use cases and build a culture of continuous improvement. At this stage, there is also the opportunity to build out additional platform functionality and more ambitious use cases across the broader journey, as the initial use cases transition into business as usual (BAU).

Examples of Data-Centric Plays in Commercial Insurance

CFC: An MGA for commercial insurance, CFC utilizes automated underwriting and data enrichment, offering rapid product development and instant binding quotes through extensive data connectivity with brokers and underwriters.
Ki: Ki Insurance, a Lloyd’s syndicate developed in collaboration with Brit, Google Cloud, and University College London, employs algorithmic risk evaluation via a broker portal, achieving £700M GWP within two years with a 31.8 percent expense ratio.
AXA: AXA XL’s Digital Ecosystem & Engagement Platform (DEEP) consolidates enterprise data, facilitating cross-selling, self-service for data professionals, and faster policy introductions, resulting in lowered expense ratios.
Cytora: Cytora, an underwriting platform, digitizes, evaluates, and intelligently triages risks, showcasing the benefits of a data-centric approach with features such as risk digitization, intelligent triaging, and performance analysis.
Hyperexponential: Hyperexponential, a pricing decision intelligence platform, handles contracts exceeding $22 billion annually.

How Publicis Sapient Can Help

In practice, very few incumbents have managed to successfully transform their underwriting process and use of data. Many insurers acknowledge the need for change but struggle to get started, while others get bogged down in large programs that fail to deliver long-term value.

Publicis Sapient is well-placed to support insurers who want to explore this transition. We have worked with insurers and other major financial services providers across the full lifecycle of transformation, putting data at the core of their businesses.

Our team of strategy, data, and customer experience experts can help insurers navigate a clear path for transformation. We can rapidly assess the opportunities available, supported by customer research and technology vendor insight, articulate a clear vision and strategy for change, and build functioning end-to-end proofs of concept in a matter of months.

Connect with Us:

DAN COLE
Senior Managing Director, EMEA & APAC
Daniel.Cole@publicissapient.com
ANDREW TAN
Principal, Insurance
Andrew.Tan@publicissapient.com
BEN RUDDLE
Senior Principal, Insurance
Ben.Ruddle@publicissapient.com

About Publicis Sapient

Publicis Sapient is a digital transformation partner helping established organizations get digitally enabled, both in the way they work and the way they serve their customers. We help unlock value through a startup mindset and modern methods, fusing strategy, consulting, and customer experience with agile engineering and problem-solving creativity. As digital pioneers with 20,000 people and 53 offices around the globe, our experience spanning technology, data sciences, consulting, and customer obsession—combined with our culture of curiosity and relentlessness—enables us to accelerate our clients’ businesses through designing the products and services their customers truly value.

Publicis Sapient is the digital business transformation hub of Publicis Groupe. For more information, visit publicissapient.com