Cohesion Lab
Menu

Services

Curate. Enhance. Teach.

Three things we do: organize your collection, explain what's in it, and turn it into teaching material.

Tier 1

Curate

Easy to find. Built to last. Ready to share.

This is the starting work. We sit down with you, sort out how each item should be described, then save the answers in formats other systems can also read. Your records are built to keep working long after this website, your current software, or us — in open formats you can take anywhere.

Deliverables

  • Records other museum and library systems can read
  • A simple display page — public or private, your call
  • Files saved in open, widely-supported formats chosen to stay readable for the long term

Shapes of work

Example scenario

An artist's estate untangles mixed-media ephemera

Audience
Executor of a mid-career artist's estate
Artifact
Sketches, letters, negatives, studio tools, and decades of show postcards

The estate needs a defensible catalog for insurance, loans, and a pending retrospective — with enough rigor to support scholarship but enough privacy to keep some material unreleased for now.

An engagement of this shape builds tiered access: a public-facing subset for the retrospective and scholarly requests, a private catalog for the family, and an embargoed layer for materials the executor wants held for a decade. Each object carries its access tier in the metadata, so changes are a flag flip rather than a migration. The insurance appraisal can run off the same dataset.

Example scenario

A family genealogy with divergent narratives

Audience
Family historian or independent researcher
Artifact
Hundreds of records across pensions, conscription notes, letters, and photographs

A family historian inherits records that trace several generations after abolition — enough material to complicate the single-narrative version of the story, but no tool for holding the complication together.

A project like this produces a private, searchable catalog with standards-based metadata — people, places, events, source documents — plus a network visualization that lets the historian and their family follow how individual choices (a conscription, a walk, a pension claim) shape generational outcomes. Provenance is recorded for each record; source scans are linked; exports are a single click.

Example scenario

A small historical society digs out a century of photographs

Audience
Volunteer-run historical society
Artifact
Photographs, letters, and ephemera across a dozen accession boxes

Three generations of volunteers maintain a ledger, a shoebox of negatives, and a shared conviction that the records will eventually "get done." A new director inherits all of it and no system.

A tier-1 engagement here catalogs the boxes across a few visits, establishes provenance notes object-by-object where the ledger is incomplete, and stands up a private online catalog the volunteers can continue updating on their own. The ledger becomes machine-readable and picks up its own backup. A first public exhibit can run off the catalog within months.

Tier 2

Enhance

A collection that explains itself.

Knowing what you have is not the same as understanding it. We add the part that explains — short notes on each item, maps, timelines, and views that show how items connect — in your collection’s own voice. We build to WCAG 2.2 AA and run an accessibility pass — automated checks plus a manual keyboard and screen-reader review — so the collection opens up rather than locking people out.

Deliverables

  • Short notes on each item, plus background and links to related sources
  • Maps, timelines, and relationship views built from your records
  • An accessibility pass — keyboard, screen readers, captions, and plain-text alternates — built to WCAG 2.2 AA

Shapes of work

Example scenario

A family archive makes decades of handwritten recipes searchable

Audience
Family archivist, three-generation collection
Artifact
Hundreds of handwritten recipe cards, annotated photographs, a shoebox of letters

The goal is to preserve the handwriting — the food stains, the margin notes, the recipe cards that are themselves objects — while making the contents searchable by ingredient, occasion, and relative.

An engagement like this layers OCR and selective hand transcription over the image archive, keeping the scanned cards as the authoritative record and treating transcribed text as a searchable index. Annotations (who wrote it, who it was for, when it was made) are queryable; the originals remain the primary objects. For a privacy-forward build of this kind, the search runs locally and no cloud processing touches the family material.

Example scenario

A library links oral-history transcripts back to its object catalog

Audience
University library with an existing oral-history program
Artifact
Hundreds of hours of transcribed interviews; thousands of objects in the main catalog

Transcripts and objects have been curated by different teams with different vocabularies. Researchers keep hitting dead ends because the two datasets don't speak to each other.

A tier-2 engagement here builds a linked-data layer that maps named entities in the transcripts (people, places, events) to catalog records, with qualified references documenting the evidence for each link. A researcher reading an object record sees every transcript passage that mentions it, with timestamps; a researcher reading a transcript sees the objects that relate to what’s being discussed.

Example scenario

A regional museum adds a public-facing timeline

Audience
Small municipal museum with one paid staff member
Artifact
A few thousand catalogued objects, strong records, no online presence

The records are solid. The online catalog is a static list. A visitor can search but can't understand.

The work adds an interpretive layer over the existing catalog: a timeline view, a few relationship maps (people, places, events), and object-level context cards with source citations. Uncertainty is preserved — where records disagree, the display says so, and the underlying records remain the source of truth. Screen-reader paths and plain-language alternates are reviewed in the same pass as the visual design.

Tier 3

Teach

In development

Built for classrooms, members, and the public.

Teach is in development — a planned service we are designing with our curriculum lead, not yet a delivered offering. A collection becomes a classroom when someone can pick it up and teach with it, so we are building Teach around lesson paths designed for your audience — K-12, adult learners, or a members’ program — with grouped item sets, prompts, and grading guides a teacher can use directly. We build to keep it working past year one. If a curriculum project is on your horizon, tell us — we will be candid about where Teach stands before we scope anything.

Deliverables

  • Lesson paths designed for your audience (K-12, college, or adult learners)
  • Sets of items grouped for a lesson, with discussion prompts and grading guides
  • A teacher's guide and a plan for keeping it all up to date

Shapes of work

Example scenario

A private foundation builds a twelve-week seminar from its archive

Audience
Cultural foundation with an existing membership program
Artifact
Mixed-media archive (manuscripts, audio, photographs) with uneven catalog depth

The foundation wants its archive to support a structured adult-learner seminar without hiring a full-time educator.

An engagement like this co-designs a few twelve-week pathways through the archive, each with object sets, readings, discussion prompts, and a rubric. The curation deepens the catalog alongside the curriculum — each pathway requires better provenance and metadata, and both improve in the same engagement. A maintenance plan hands the next year of updates to the foundation’s own staff.

Example scenario

A small museum ships a K–12 classroom module

Audience
Education director at a small regional museum
Artifact
A roughly thousand-object catalog centered on a single 19th-century industry

The museum wants a teachable unit that works for local middle-school field trips AND for remote classrooms that can't visit.

An engagement of this shape builds a small module (roughly three lessons) with object sets, primary-source activities, and assessment rubrics — aligned to two state standards without being locked to them. A remote version uses the same pathways through the public catalog, with educator-guided prompts that don’t require a tour. The director can update lessons themselves after handoff.

Example scenario

A university special collection supports a graduate seminar

Audience
Special-collections librarian and seminar instructor
Artifact
Manuscripts, broadsides, and ephemera from a regional 19th-century press

The seminar is redesigning itself around the collection, and the collection needs interpretive pathways the instructor can adapt each term.

A tier-3 engagement here produces a few week-by-week pathways through the collection — tied to the seminar’s reading list, with object sets, discussion prompts, and a rubric the instructor can edit. The pathways live beside the catalog, so as the collection grows, the seminar’s options grow with it. Research output from the seminar feeds back into the catalog as new annotations.

Available across tiers

Who sees what.

Most collections aren't all public or all private. We set the rules with you — item by item, group by group, even by date — so some things are open to everyone and others stay with the people who should see them. For collections that come from Indigenous communities, those communities set their own rules[5] — and we take on that work only in partnership with the community. Until we have an established Indigenous partnership, we decline it. You can add this to any service.

What it costs

What an engagement costs.

We price the work, not your collection's importance. What an engagement costs depends on a handful of practical drivers — how much material there is, the condition and formats it's in, how much description and interpretation you want, and how the results need to be delivered and access-controlled.

Rather than publish list prices that can't fit work this varied, we scope each engagement and give you a single fixed price in a written proposal. The path is short: tell us about your collection through the contact form, we follow up with a brief discovery conversation, and you receive a fixed-price proposal. If we aren't the right fit, we'll tell you early and point you toward someone who is.

Want to lower the cost? Preparation you do yourself — organizing files, supplying a basic inventory of what's in the collection — reduces the effort the engagement takes, and we'll tell you exactly where it helps.

Not ready to engage but starting work yourself? Our free starter checklist for family archives is the kind of work you can do before hiring anyone — and it stays useful if and when you do.

Sources

The plain-language notes above point here. Each entry names the technical term, a one-line meaning, and the original source.

  1. FAIR. A checklist for keeping records easy to find, open, share, and reuse over time. Stands for Findable, Accessible, Interoperable, Reusable. Used by museums, libraries, and research data centers worldwide. Wilkinson, M.D. et al. (2016). Scientific Data 3:160018. https://doi.org/10.1038/sdata.2016.18
  2. Shared formats for records. Common ways to describe items so other museum and library systems can read them. The three we use most are Dublin Core (a short, plain set of fields), CIDOC-CRM (a museum-wide map of how things relate), and schema.org (the same vocabulary search engines use). https://www.dublincore.org/specifications/dublin-core/dcmi-terms/
  3. IIIF. A shared way to publish high-resolution images so anyone can zoom in, compare side-by-side, and link to a specific view. Used by many of the world's largest libraries and museums. https://iiif.io/
  4. NDSA Levels of Digital Preservation. A plain checklist for keeping digital files safe for decades — covering things like backup copies, file formats, and how often you check the files are still good. https://ndsa.org/publications/levels-of-digital-preservation/
  5. Indigenous data sovereignty. When a collection comes from an Indigenous community, that community sets the rules for how items are shown, shared, and used. We defer to the Local Contexts framework via credentialed partners. Traditional Knowledge (TK) and Biocultural (BC) Labels are authored and applied by the community itself — never by us — and we decline Indigenous-collection work until we have an established Indigenous partnership. https://localcontexts.org/

Not sure which tier fits?

Most collections don't arrive tier-labeled. Describe what you have, and we'll sketch a scope and a sequence — no cost, no pressure.