• Data Strategy

29 Jun, 2022

Data: Our most valuable resource, without accounting monetary value

Do you know the monetary value of your data? If your answer is no, you are not alone.

Treating data as an asset is not a new topic, research shows that many executives are trying to understand the financial value of their data (like any other asset within their organisation). 

So why is data not captured on balance sheets and how can we change this?

Balance sheets and Profit & Loss relate to accounting artefacts. Conventional accounting systems and methods do not currently assign a monetary value for data. This is because there is no consensus on how to do it for financial reporting.  Yet, without a measure of the value of data, leaders lack the financial information to make the most effective decisions about where investments should be made. 

For example, if a company were to spend an additional 2% of its budget to improve the quality of its data, that decision would only be reflected on the books as a cost. The improved quality of the data would not be measured or accounted for anywhere.[1] How would organisations calculate whether that investment has any returns?

Clearly there is a need to valuate data, and it to be officially documented in order to have a reference point to aid strategic decision making. As data professionals, we cannot wait for accounting standards to catch up with us, we must find alternative ways to place value on data. Some organisations are setting up Data Valuation Offices, thought leadership and methods and frameworks to approach the lack of official process.

At Chaucer we recommend taking a fundamental sound consistent and transparent framework to evaluate the value of data assets. We have collected many pieces of research complemented, by our extensive experience, to develop the following approach.

The approach to valuate data

Our valuation approach combines traditional methods of valuing assets with new innovative data driven approaches

Initial Steps to Valuing Data

  1. Data Discovery: Discover what data exists within the organisation’s ecosystem
  2. Data Assessment: Assess the data sets, where they reside, their quality and ownership
  3. Data Prioritisation: Identify and group valuable data assets or priority domains
  4. Data Valuation: Apply a financial/accounting valuation method

Developing the right framework

Discover what data exists within the organisation’s ecosystem

As a first step in the framework, we must understand the data available, and systems used to create, analyse, and process information. This vision of the data landscape will provide the overview necessary to assess the data and determine its value-proposition for the business. Depending on an organisation’s maturity, existing artefacts may range from data asset registers, data catalogues, spreadsheets, to nothing at all - in that case a data discovery exercise will need to be completed.

Assess the Data

With an overview of the data’s landscape at an enterprise level or within a particular domain, we can evaluate each data-assets' purpose, activity, and create an understanding of what business value the data asset brings.

Looking at 12 different key data-asset categories (such as profitability, operational, compliance, or data storage, management, and governance, quality), we apply weighted scores to indicate the most opportune or high-risk areas to focus on within the organisation's business model.

A snippet of a data assessment template is shown below. The purpose of this template is to provide a tool for collecting, tabulating and summarising the inherent value of a particular unit of data (Ray, 2021).

Identify & group perceived valuable data

Not all data within an organisation has value therefore it is important to select the data assets that areuseful and supports improving business value. This can be driven from a:

(i) top-down approach by defining use cases within a data strategy, critical reports, or

(ii) grouping assets from the assessment done in stage 2.

Do not ‘boil the ocean’. Large organisations could have 1000’s of data assets therefore the task of valuing these assets would potentially lose its value. We need to be able to group these into logical units of relatable data assets before applying a valuation technique to that group.

Apply a data valuation method

There are many different approaches that can be taken to assess value. Applying a data valuation method will depend on the data valuation goals the organisation has set and alignment to overall strategy.

Doug Laney is the Data & AI Strategy Innovation Fellow at West Monroe and is also the bestselling author of Infonomics: How to Monetise, Manage, and Measure Information for Competitive Advantage. In his book he describes two types of valuation approaches. Foundational and Financial. If an organisation’s priority is in optimising their information management practices, they may track the foundational measure of information value. If, on the other hand, they want to understand the financial impact of information on the business, they can leverage the financial measures of information value.[2]

Foundational models: tend to be more suited for internal valuation methods that are not subject to financial accounting principles. They are also more subjective and can be theoretical causing inconsistencies in definitions \ calculations across the business.

Intrinsic Value of Information (IVI) - Measures data quality of data assets including correctness, completeness, and exclusivity of data

Business Value of Information (BVI) - Measures how fit the data is for specific business purposes (e.g., initiative X requires 80% accurate data that is updated weekly)

Performance Value of Information (PVI) - Measures how the usage of the data effects key business drivers/KPIs - often using a control group study

Financial models: These valuation methods are more aligned to those that will appear on a balance sheet and the business outcomes we mentioned earlier.

Cost Value of Information (CVI) - Measures the cost to produce and store the data, the cost to replace it, or the impact on cash flows if it was lost

Economic Value of Information (EVI) - Measures the expected cash flows, returns, or savings from the usage of the data

Market Value of Information (MVI) - Measures the actual or estimated value the data would be traded for in the data marketplace

Expected Value of Perfect Information (EVPI) – Measures the value of the information you don’t have

Which data valuation method to use?

Depending on the objectives, we feel that these methods and formulas (either foundation or financial) should not be used in isolation and can be combined to gain a broader perspective. However, the main challenge is in the application themselves, even more so for the foundation models. For example, in calculating the Intrinsic Value of Information (IVI) there are 4 variables that need to be calculated.

  • Validity: The percent of records with correct values
  • Scarcity: An estimate of the percent of other organisations who don’t have this data.
  • Coverage: The number of records in the dataset as a percentage of the total universe of potential records
  • Useful Life: The number of periods (months, for example) that each record can reasonably be used or is relevant

How validity is calculated is likely to vary and is open to interpretation unless there is a standard way agreed within the business or even better, by industry. Calculating a value for Scarcity is even more subjective. Coverage and Useful Life can be calculated, but still need a standard way to be assessed to be comparable. 

The financial models, however, seem to be more data driven. Take for example the Economic Value of Information (EVI) which measure deals directly with revenue and is a very powerful and tangible measure of the value of data.[3] The variables required are:

Revenue(with) — Revenue generated with the data

Revenue(without) — Revenue generated without the data

Cost of Data — The cost to acquire, administer, and operationalise the data in relevant business processes

T — The usable life of any datum

t — The period over which the experiment was executed

Cost of data and Revenues are well understood metrics, even if you may need to run experimentation and possibly even a Monte Carlo Simulation to determine future values. The advantage with the financial models is that the variables are easily understood, comparable internally across domains and therefore become an internal currency or measure which aids in the prioritisation of initiatives.

In addition to Doug Laney’s methods above there are many other financial valuation methods such as the Cost-based approach or Market approach.

In a nutshell…

In the accounting world, data is not seen as an asset due to lack of standards to calculate it hence not being represented on a balance sheet. This does not mean that data does not have an intrinsic economic value; it just means that we must find alternative ways to calculate what the value is and how it relates to the performance and value of an organisation. 

From our analysis, the financial methods of determining value seem to be more practicable, credible, easier to calculate, and have fewer assumptions and subjectivity in their calculations. Making Data Valuation part of data strategy should be essential to help support gathering buy-in and business case development.

Developing an analytical approach to determine the value of data can quickly turn into a complex exercise if a structured framework and focused objectives are not followed. These are our recommendations:

  • Define your valuation objectives and focus on a distinct set of related data. This could be as simple as assessing the data that contributes to a report or focusing on data from a single business domain.
  • Follow a structured approach to systematically work through the data and understand which data assets support decision making and value generation.
  • Do not treat this as a one-time exercise. Organisations should develop a process to continually review and enhance data valuations.
  • Create a data valuation document to be used as a reference point, to then communicate and collaborate with the business. It will help identify data initiatives to be prioritised in a roadmap based on data valuation document.

Conducting a Data Valuation exercise will provide the necessary insight that organisations need to understand the value of their data, in a language that most people understand – money! Having that monetary value of data will help identify the data assets that they need to nurture, invest in, and protect vs data that they just need to manage. Having this understanding will support executive decision making, prioritise data programs, establish IT priorities, support business valuation and even, dare I say it, maybe one day, become a line item on a balance sheet.

Author: Justice Allotey – xTech Data Strategy Lead

Contributor: Elodie De Fontenay – UK & US xTech Lead

Contributor: Natan Ostro - xTech Data Consultant


(n.d.). Retrieved from https://www.ifrs.org/issued-st...

(n.d.). Retrieved from https://www.cpajournal.com/201...

(n.d.). Retrieved from https://www.mckinsey.com/busin...

Benyayer, L. D. (n.d.). 4-methods-to-measure-the-value-of-data/. Retrieved from louisdavidbenyayer.com: https://www.louisdavidbenyayer...

how-to-measure-the-value-of-data. (2021, 03 27). Retrieved from showmethedata.blog: https://showmethedata.blog/how...

Ray, D. E. (2021). Valuing Data - An Open Framework . Boca Raton: CRC Press.

[1] (Ray, 2021)

[2] How to Measure the Value of Data — 7 Ways to Inform Your Data Strategy - Show Me The Data

[3] How to Measure the Value of Data — 7 Ways to Inform Your Data Strategy - Show Me The Data

Blog 11 Feb, 2022

Data Strategy, Data, RPA, AI, Robotics, Machine Learning, CX

A Machine Learning primer

11 Feb, 2022

Smitha Dunwell

Managing Principal

Blog 28 Sep, 2021

Data Strategy, Data Science & Analytics, Data Visualisation, Data, Data Ethics, Culture

Finding our equilibrium workplace

28 Sep, 2021

Jill Dawson

Head of Marketing

Blog 23 Jul, 2021

Data Strategy, Data Science & Analytics, Data Visualisation, Data, Data Ethics

Five risks of getting ‘return to work’ wrong and how to avoid them

23 Jul, 2021

Elodie De Fontenay

Insight Partner – Data & AI