Before a library can classify, catalogue, lend, preserve or recommend anything, something has to enter the collection.
That entrance is not a single act. It is a controlled chain of decisions.
A librarian identifies a need. A title is evaluated. A budget is checked. A supplier is chosen. Terms are reviewed. An order is placed. Material arrives—or access is activated. The item is verified, recorded, processed and handed to the next part of the library system.
This article is part of eduKateSG’s How X Works programme and the How a Library Works series.
The shortest useful answer
Library acquisition works by converting a collection need into a controlled resource that the library can legally, financially and operationally manage.
The word controlled matters.
A book donated in a box is not automatically a library resource. An ebook paid for is not automatically usable by every patron. A journal subscription is not the same as permanent ownership. A digital file copied onto a server is not properly acquired if rights, provenance, format or preservation obligations are unknown.
Acquisition is complete only when the institution understands what it has obtained, under what conditions, at what cost, for whom, and with what continuing responsibilities.
Selection comes before acquisition
Libraries cannot acquire everything, so they need rules for deciding what deserves to enter.
Selection may be shaped by:
- the mission of the library;
- age and needs of users;
- curriculum or research programmes;
- current and expected demand;
- quality and authority;
- language and cultural coverage;
- existing collection strengths and gaps;
- price;
- storage and preservation capacity;
- format;
- licensing conditions;
- local relevance;
- scarcity or archival significance.
Selection is therefore a prioritisation problem. The library is allocating finite attention, money and future maintenance capacity across a much larger universe of possible resources.
Acquisition begins with a collection policy
A mature library does not want every purchase decision to depend entirely on the preference of one person on one day.
Collection-development policies create continuity. They define the communities served, subject priorities, preferred depth, formats, exclusions, donation rules, replacement principles and responsibilities for review.
The policy acts like a boundary condition. It does not make every decision automatically, but it narrows the space of acceptable decisions and makes those decisions more explainable.
The main acquisition routes
Libraries obtain resources through several routes, each creating different rights and obligations.
Purchase
The library buys a physical or digital resource from a publisher, bookseller, distributor, platform or other supplier. For physical material, ownership may transfer to the library. For digital material, payment may instead purchase a licence rather than ownership.
Subscription
Journals, databases, newspapers, streaming services and research platforms are often supplied through time-limited subscriptions. Continued access depends on continued payment unless the contract includes perpetual-access rights for previously paid content.
Donation or gift
Individuals, authors, institutions or estates may donate material. The important question is not whether the material is free, but whether accepting it creates more value than cost. Donations still require evaluation, processing, storage and long-term stewardship.
Legal deposit
In some jurisdictions, designated libraries receive publications under legal-deposit arrangements. This supports national documentary memory by creating systematic routes through which published output enters preservation collections.
Exchange and institutional transfer
Libraries and organisations may exchange publications or transfer records and collections internally. Universities may receive departmental archives; government libraries may receive agency publications; museums and libraries may transfer custody according to collecting responsibilities.
Digitisation and born-digital deposit
A library may create digital surrogates of physical holdings or acquire born-digital material directly. This introduces technical requirements such as file validation, checksums, storage planning and rights documentation.
Ordering is a supply-chain problem
Once a title is selected, acquisition becomes partly logistical.
The library must identify an edition, determine availability, choose a vendor, place an order, track fulfilment, receive the item, match it to the purchase record and resolve discrepancies.
At scale, libraries may work with approval plans, standing orders, subscription agents, aggregators and vendor platforms. These mechanisms reduce the transaction cost of acquiring thousands of resources individually.
The acquisition department is therefore a bridge between intellectual collection strategy and operational procurement.
Why editions matter at the point of purchase
Buying the correct title but the wrong edition can create a real service failure.
A textbook edition may align with a different syllabus. A legal text may omit recent amendments. A medical reference may be superseded. A translation may be intended for a different audience. A paperback and hardback may differ only in format—or may contain revised material.
Identifiers such as ISBNs help reduce this ambiguity, but human review is still necessary when bibliographic data is incomplete or inconsistent.
Receiving closes the order loop
When physical material arrives, the library checks whether reality matches the order.
Is it the correct title? Correct edition? Correct number of copies? Is the item damaged? Was an invoice duplicated? Did a multi-volume set arrive complete? Is a supplement missing?
Receiving is a reconciliation stage. It prevents procurement data from drifting away from the physical collection.
Once verified, the item can be accessioned or otherwise registered and passed into cataloguing, classification, physical processing and shelving.
Accessioning: recording the fact of entry
Accessioning formally records that material has entered a collection or custody system.
Practices differ among libraries and material types, but the underlying purpose is stable: establish an auditable transition from external object to institutional responsibility.
For rare books, archives or special collections, accession data can be particularly important because provenance, source, donor terms and chain of custody may later carry research significance.
The budget is not just a spending limit
Acquisition budgets force a library to reveal its priorities.
A fixed sum must be distributed across subjects, formats, age groups, subscriptions, replacements, emerging needs and long-term commitments. Some resources are one-time purchases. Others create recurring costs that compound year after year.
This makes acquisition a portfolio problem.
A library that spends too heavily on expensive recurring subscriptions can lose flexibility. A library that buys only popular current titles may weaken long-term research depth. A library that preserves every historical commitment may leave too little money for new fields.
Good acquisition therefore balances present demand with future usefulness.
Digital acquisition is often licensing, not ownership
This is one of the most important differences between print and digital collections.
A library that buys a printed book usually receives a physical object it can keep subject to law and policy. A library that pays for an ebook, journal package or database may receive permission to access content under contractual conditions.
Those conditions can govern:
- who counts as an authorised user;
- whether off-site access is allowed;
- how many people may use a resource simultaneously;
- whether chapters can be downloaded or printed;
- whether content can be used in course systems;
- whether text and data mining is permitted;
- what happens when the subscription ends;
- whether archival access survives cancellation.
The acquisition decision must therefore evaluate both content and contract.
Perpetual access: what does “we bought it” really mean?
Digital collections expose a subtle question: if a library pays for access for ten years and then cancels, what remains?
Some agreements provide continuing access to content paid for during the subscription period. Others do not. Access may depend on a publisher platform that could later change ownership or technical systems.
This means preservation and acquisition become entangled. A library may need contractual rights, technical infrastructure and cooperative preservation services to ensure that purchased scholarly content remains usable over time.
Donations are not free
A donation may cost nothing to purchase and still be expensive to accept.
Staff must evaluate it. Boxes must be transported. Items may need cleaning or conservation. Records must be created. Shelving space must be found. Duplicate material must be reviewed. Restrictions may have to be documented.
For this reason, libraries often reserve the right to decline gifts or to accept them without promising permanent retention.
The question is not whether a donor values the material. The question is whether the material fits the collection’s mission strongly enough to justify the library’s continuing responsibility.
Demand-driven acquisition
Digital systems make it possible to delay some purchase decisions until users demonstrate demand.
Under demand-driven or patron-driven acquisition models, large pools of records may appear discoverable even though the library has not permanently purchased every item. A user action can trigger a short-term loan or purchase according to pre-set rules.
This changes acquisition from a purely anticipatory model to a partially responsive one.
The advantage is efficiency: money follows observed use. The risk is that immediate demand can dominate less visible long-term collection needs.
Evidence-based collection development
Libraries increasingly combine professional judgement with data.
Useful signals include circulation counts, holds, failed searches, database usage, cost per use, interlibrary-loan requests, course reading lists, citation patterns, publication trends and user requests.
No one measure is sufficient. A rare local-history item may have low circulation but high preservation value. A heavily used textbook may justify many copies but have little long-term archival importance. A costly research database may serve a small group whose work is central to the institution’s mission.
Acquisition therefore requires both measurement and interpretation.
Replacement is acquisition too
Collections do not grow only by adding new subjects. They also maintain capability by replacing what has been lost, damaged or superseded.
A replacement decision asks whether the same item should be reacquired, whether a newer edition is better, whether a digital version should replace print, or whether demand has fallen enough that no replacement is necessary.
This is maintenance capital for knowledge infrastructure: spending to preserve an existing service rather than expand it.
Acquisition and preservation meet at the beginning
Preservation should not be treated as something that begins decades after acquisition.
The moment a library accepts a fragile manuscript, an obsolete digital format, magnetic media, a website archive or a proprietary database, it is already accepting preservation risk.
Good acquisition therefore asks: can we actually keep this usable?
For digital material, this may mean checking file formats, storage requirements, fixity, metadata completeness and transfer rights before acceptance. For physical special collections, it may mean considering conservation needs, environmental conditions and secure storage.
Acquisition as risk management
Every acquisition carries risk.
- The resource may be underused.
- The supplier may fail.
- The licence may become unaffordable.
- The format may become obsolete.
- The content may duplicate existing access.
- A donation may carry difficult restrictions.
- A popular resource may become outdated quickly.
- A rare resource may require expensive preservation.
The library cannot eliminate these uncertainties. It can make them explicit, compare them against mission value, and avoid taking obligations it cannot sustain.
What AI changes in acquisition
AI can help libraries scan publication markets, cluster user requests, identify collection gaps, compare vendor offers, extract licence clauses and detect emerging subject demand.
But acquisition remains a governance decision.
A model can estimate relevance; it cannot independently decide what a community should remember. It can predict demand; it cannot define the institution’s mission. It can compare costs; it cannot settle every question of cultural significance, fairness, preservation or intellectual breadth.
The best use of AI is therefore to strengthen evidence and reduce repetitive work while leaving accountable collection decisions inside the library’s governance framework.
The complete mechanism
- The library defines its mission and collection policy.
- A need or opportunity is identified.
- Candidate resources are evaluated for fit, quality, demand, cost and sustainability.
- The appropriate acquisition route is chosen: purchase, subscription, donation, deposit, exchange, transfer or digital ingest.
- Edition, format, supplier and rights are verified.
- Budget authority is confirmed.
- The order, agreement or transfer is executed.
- Physical material is received and reconciled; digital access or files are validated.
- The resource is formally registered under library control.
- It moves into cataloguing, classification, processing and access.
- Usage, cost, condition and continuing relevance feed back into future acquisition decisions.
That is how library acquisition works.
It is the moment where a library turns judgement into responsibility: deciding not only what should be collected, but what the institution is prepared to identify, fund, maintain, preserve and keep reachable for someone who may need it later.