Can you tell what your learning content is worth? Or how much you spend on it?

Do you know how many courses sit in the library. Or the return on it?

That last point is harder to answer, and in an era of tightening budgets, it is increasingly the question that matters.

This guide walks through a practical framework for calculating the ROI of your learning content estate, and the data you need to make that calculation meaningful.

Content typically makes up 42% of total L&D spending once you factor in licensing, production and curation time. Our own analysis of content estates with 10,000+ employees found an average overlap of at least 20% across all formats, rising to 31% for courses alone. That means a fifth or more of most content budgets is spent on material that already exists elsewhere in the estate, before anyone has even asked whether it closes a real skills gap.

Why content ROI is hard to measurehere the costs accumulate

The standard approach is to track completion rates and learner satisfaction scores. Neither tells you much about value. A course with 80% completion may be mandatory and resented. A course with 10% completion may be reaching exactly the people who need it.

The real question is not how much learning happened. It is whether the learning that happened addressed a genuine skills gap, and whether that gap closure has a measurable business outcome attached to it.

To answer that, you need three things: a clear picture of your content estate, a skills framework your content is mapped to, and workforce data that tells you where the gaps are.

Step one: Take stock of what you have

Before you can measure ROI, you need an accurate inventory. For most large enterprises this is harder than it sounds. Content lives across multiple systems: an LMS, an LXP, a content library or several, SharePoint, Teams channels, and various legacy platforms. Assets are often duplicated, mislabelled, or no longer relevant to the business.

A content audit should capture: the total number of assets, the format and source of each, when it was last updated, its usage data where available, and whether it maps to a current skills priority.

Signal Quality is built to automate exactly this audit. It scores every asset in your estate continuously, checking each one for relevance, quality, freshness and duplication. It shows you what overlaps, what has gone stale, and what is actually being used. That turns Step one from a multi-month manual project into an always current view of the estate, one that keeps working as content is added and retired.

Step two: Map content to skills and business priorities

Once you have an inventory, the next step is to connect content to skills. This means taking your skills framework, whether that is built on a proprietary framework or job architecture from your HRIS, and identifying which assets address which skills.

This is where manual curation reaches its limits. At enterprise scale, mapping thousands of assets to hundreds of skills nodes requires automation. Without it, the process either does not happen or produces results that are out of date within months.

Step three: Identify the waste

Content ROI has two sides. The numerator is the value generated by content that is used, relevant and effective. The denominator includes all content spend, including the spend on content nobody uses.

In most enterprise estates, duplication is significant. Multiple assets covering the same topic, from different vendors, purchased in different years, sitting in different systems. If nobody is deduplicating, you are paying for the same material several times over.

The AstraZeneca case study illustrates this clearly. By running a structured content audit, AstraZeneca identified a 41% saving on learning content spend. The saving was not from cutting programmes. It was from removing duplication and redirecting budget to content that was actually filling skills gaps.

Step four: Calculate the number

A simple ROI model for a content estate looks like this:

Take your total annual content spend (licences, production, curation time, platform costs). Subtract the cost of content identified as unused or duplicated. Add the estimated value of skills gap closure, using business metrics like time-to-competency, reduced error rates, or avoided external hiring costs where you have them. Divide net benefit by total cost.

What good looks like

Organisations that have invested in learning infrastructure report significant improvements in content ROI once they have a clear picture of their estate. The reason is straightforward: you cannot optimise what you cannot see. Getting the data in one place, mapped to skills and business priorities, is what makes the calculation possible in the first place.

Filtered Intelligence connects your content estate, skills data and learning systems into a single structured layer. Signal Quality scores every asset continuously for relevance, quality and duplication, giving L&D leaders the visibility they need to make these decisions with confidence.