Usage Bibliometrics (Downloads and COUNTER)
Also known as: Download Metrics, Usage Factor Analysis, Usage-Based Impact Metrics, COUNTER Usage Analysis
Usage bibliometrics measures the impact of scholarly works from how often they are downloaded and viewed rather than how often they are cited. Drawing on server and publisher logs standardized through the COUNTER code of practice, it turns raw access events into impact indicators such as the usage factor. The MESUR project led by Johan Bollen and Herbert Van de Sompel was pivotal: their 2008 work demonstrated usage-based impact metrics built from large-scale usage logs, and their 2009 principal component analysis of thirty-nine impact measures showed that scientific impact is multidimensional, with usage metrics occupying a distinct region of the space from citation metrics. Usage signals accrue almost immediately and reflect a far larger readership than the subset of authors who eventually cite, making them an early and broad complement to citation analysis, provided the logs are carefully standardized.
Key highlights
- Provides an immediate impact signal that appears as soon as a work is accessed, well before citations.
- Reflects the entire readership, including non-citing audiences that citation analysis cannot see.
- Standardized through COUNTER, enabling comparable usage statistics across platforms and publishers.
- Empirically shown to occupy a distinct dimension of impact, complementing citation metrics.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use usage bibliometrics when you need an immediate signal of attention, when you want to measure engagement from the full readership rather than just citing authors, or when you are evaluating resources such as journals, repositories, or datasets where libraries track access. It is appropriate where standardized usage logs are available, COUNTER-compliant statistics in particular, and where the volume of accesses is large enough to be informative. It is valuable for collection management, for early assessment of recent outputs, and for studying the multidimensional nature of impact. It is less appropriate where logs cannot be cleaned to a common standard, where usage is dominated by automated traffic, or when the question specifically concerns scholarly influence as expressed through citation. Privacy and gaming concerns also constrain how usage data can be collected and reported.
Strengths & limitations
- Provides an immediate impact signal that appears as soon as a work is accessed, well before citations.
- Reflects the entire readership, including non-citing audiences that citation analysis cannot see.
- Standardized through COUNTER, enabling comparable usage statistics across platforms and publishers.
- Empirically shown to occupy a distinct dimension of impact, complementing citation metrics.
- Raw logs are contaminated by robots and double clicks and require careful, sometimes imperfect, standardization.
- Usage is easily inflated or gamed, and a download does not imply reading or scholarly use.
- Coverage depends on access through tracked platforms, missing copies obtained elsewhere.
- Privacy constraints and proprietary log ownership limit the availability and granularity of usage data.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
What is COUNTER and why does it matter for usage metrics?
COUNTER, short for Counting Online Usage of Networked Electronic Resources, is a code of practice that specifies how usage events should be filtered and counted, for example by removing robot traffic and collapsing rapid double clicks, so that statistics from different publishers and platforms are comparable. Without such standardization, raw server logs differ so much in logging practice and bot contamination that their counts cannot be compared. COUNTER compliance is therefore the foundation of usage bibliometrics: it turns idiosyncratic logs into trustworthy, interoperable usage statistics that libraries and analysts can rely on.
How do usage metrics differ from citation metrics?
They measure different things and on different timescales. Citations record formal scholarly influence and accrue over years; usage records access and accrues immediately, reflecting the whole readership rather than just citing authors. Bollen and colleagues' principal component analysis of thirty-nine measures showed usage and citation metrics cluster in separate dimensions, meaning impact is genuinely multidimensional. The two are moderately correlated, so usage is informative, but it is not a substitute for citation; the value of usage bibliometrics lies precisely in capturing an early, broader facet of impact that citations miss.
Can usage statistics be trusted given that downloads can be gamed?
They can be trusted only after careful cleaning and with appropriate caution. Raw logs are vulnerable to robot traffic, accidental repeat clicks, and deliberate inflation, which is why COUNTER filtering and robot exclusion are mandatory before any analysis. Even then, a download signals access, not reading or endorsement, so usage metrics should be interpreted as attention indicators and reported alongside other measures. Used this way, with standardization and as one dimension among several, usage statistics are a reliable and valuable complement; treated as a sole, raw measure of quality, they are easily misleading.
Sources
- 1.Bollen, J., Van de Sompel, H., Hagberg, A., & Chute, R. (2009). A Principal Component Analysis of 39 Scientific Impact Measures. PLoS ONE, 4(6), e6022.
- 2.Bollen, J., Van de Sompel, H., & Rodriguez, M. A. (2008). Towards usage-based impact metrics: first results from the MESUR project. Proceedings of the 8th ACM/IEEE-CS Joint Conference on Digital Libraries (JCDL), 231-240.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Usage Bibliometrics (Downloads and COUNTER). ScholarGate. https://scholargate.app/bibliometrics/usage-bibliometrics