Architecture is applied precedent. A designer cannot make a load-bearing decision in isolation, cannot choose a cladding system without understanding how it behaves in this specific climate, and cannot detail a transition without seeing how others solved it before.
This is why architects keep libraries. Not of finished work from their own office, but of every building that taught them something about how materials age, how ventilation works in a monsoon climate, what happens to a certain detail thirty years and five thousand cycles of freeze-thaw later.
Precedents are scattered: a vernacular study from a journal, a photograph from a site visit, a material sample tagged with source information, sections from an academic paper, site conditions documented from a place you worked. They live in different folders because they arrived at different times for different reasons. A reference system that can search across that whole landscape at once makes the research phase faster and keeps the best learning from every project available when you need it.
This is a day of research for a residential project in a cold climate.
Morning — Set up the project workspace
The project folder is /Projects/2026-NewHome-Vermont/ and you've created two subfolders: Precedents/ for finished buildings, and Materials/ for cladding and exterior detail studies. The workspace will add the Coii Ref database alongside them.
Open the app and start with the research questions you need answered:
- What do residential buildings do in cold climates to manage moisture at the wall?
- How do architects in Vermont handle the wall detail where the roof meets the exterior?
- What cladding holds up in a salt-spray environment?
Search for precedents you've collected on cold-climate residential:
tag:cold-climate-residential rating:>=3 type:imageYou have eighty-four images: completed houses in New Hampshire, Quebec, Norway, and upstate New York. Some are your own photographs from site visits. Some are from publications. All of them are tagged and live in your precedent archive. Sort by rating to find the strongest examples first.
10:00 am — Study wall details at the envelope transition
The critical detail is where roof meets wall. Pick your three strongest precedents and open them side by side. For the first, a 1970s farmhouse in New Hampshire:
- Standing-seam metal roof.
- Wood siding on the exterior wall.
- A clear overhang; no drip edge confusion.
Open the building's floor plans and sections in the same app to get the interior construction sequence. The washboard below the fascia acts as a capillary break. The soffit is vented. The siding overlaps far enough that water sheds outward.
Rate the roofing detail a 5 and add a note: "Aggressive overhang strategy. Thirty years, no rot." Add the image to a new canvas called "Roof-Wall Transition." You will build this board over the next few days as you research, and it will document the approach you choose.
For the second precedent, a 1990s house in Quebec:
- Metal roof again, but with a gutter system.
- Exterior insulation over the sheathing.
- No siding; the insulation is the weather barrier.
This is a different approach entirely. Rate it a 4—it is legitimate, but ice-dam risk exists because the gutter can clog. Add a note: "Continuous insulation works in this climate, but maintenance is higher." Do not rate this approach down because of one weakness; rate it fairly for what it accomplishes.
11:30 am — Search the material library for compatible cladding
Search materials you've collected that work with the thermal strategy you're considering. If you go with the overhanging roof and traditional wood siding, cladding choices expand. If you go with exterior insulation, material options are narrower:
tag:cladding,exterior-insulation type:image in:/Users/Architect/Archive/MaterialsYou find forty-two materials studies. Many are samples—photographs with the source information in the note. Some are full walls. Filter to just the ones rated as proven in cold climates:
tag:cladding,exterior-insulation,cold-climate rating:>=4Twelve remain. These are the materials that have been tested in freeze-thaw conditions. Pull the three strongest onto the "Roof-Wall Transition" canvas alongside the building precedents. A material does not stand alone; it stands in context of the wall strategy and the climate it lives in.
1:00 pm — Review the site conditions precedents
Your building site is a south-facing slope with existing trees to the north. Search precedents that face similar conditions:
tag:site-conditions,slope,sun-exposure-south type:imageYou find nineteen images. Rate the ones that handle the slope intelligently—the ones where the designer worked with grade rather than against it. Add notes about grading strategy and foundation type.
The trees are a wind break and a solar buffer in summer. Search:
tag:tree-line,wind-protection rating:>=3Fourteen images. The best one is a house you visited in New England where deciduous trees north and east create wind protection while allowing winter sun. Add a note: "Cooling in summer, solar gain in winter." Rate it a 5.
These details will not end up in the final design, but understanding how other designers handled them is the research that keeps your own decisions honest.
2:30 pm — Find material failures and learn what not to do
Precedent research is not just what worked. It is also what failed and why. Search for images tagged with failure documentation:
tag:failure,cold-climate type:image rating:<=2You have photographs of:
- Wood rot at the base of siding due to inadequate overhang.
- Ice damming at a gutter that was not sloped correctly.
- Exterior insulation that failed at the edge detail because the flashing was not specified carefully.
Rate these a 1 or 2, but add detailed notes. "This building's wall is failing because the overhang was 18 inches instead of 24, and snow accumulation against the siding caused wetting." This information is as valuable as the successful examples.
Create a tag failure-analysis-cold-climate and add it to each one. The next time you start a cold-climate project, these will be the first thing to review.
3:30 pm — Update the design board with your research conclusions
By now you have photographed or collected:
- Three roof-wall transition strategies from precedent buildings.
- Twelve cladding options that work in your climate.
- Material failure examples to avoid.
- Site condition strategies from four similar conditions.
Save this canvas as "Design Research — Vermont House" in your project folder. This is now version-controlled. If you change direction next week, you can return to this board and see exactly what was being considered.
Add notes to the canvas explaining your thinking:
- "Overhang strategy chosen because existing trees to north reduce summer cooling load; south exposure prioritizes solar gain."
- "Exterior insulation rejected due to higher maintenance risk in this climate; standing-seam cladding chosen instead."
These notes stay with the canvas. When the client asks why you chose this detail, you show them the precedents that informed it.
4:00 pm — Save a research view for the next phase
The research you are doing now will be needed again when you detail the building. Save a view:
tag:cold-climate-residential,roof-wall-transition rating:>=3 type:imageSave another:
tag:failure-analysis-cold-climate type:imageSave a third:
tag:cladding,cold-climate,proven type:imageEach saved view is a search that runs in less than a second. When you open the file six months from now during detailing, these views are waiting and will show you exactly the images you filtered for during design development.
5:00 pm — Catalog the new precedents you found today
Today you added six new precedents to the library. Photograph them, add clear names, tag them immediately:
tag:cold-climate-residential,roof-detail-study added:2026-09-11The watcher saw them land. Add ratings as you tag. Add notes while the source information is still clear in your memory. Do not wait until next week.
A month later: Detailing begins
You begin to detail the wall assembly. Open the project folder and open the "Design Research" canvas. It shows the three roof-wall approaches you were considering and the cladding studies you chose.
You still have access to every precedent building. Search by tag to pull the one you want:
tag:cold-climate-residential,roof-wall-transition,chosen rating:>=4Pull up the section detail from that building. Measure the overhang. Understand how the gutter—if there is one—is designed. Look at the flashing detail. None of it is copied; you are reading the precedent to inform your own specification for this project.
The material library shows every cladding you considered and rated. Your best choices are already marked as proven. Pull them one more time:
tag:cladding,cold-climate,proven rating:>=4Your detailer can see these images in context and understand not just that you chose this material, but why—because it survived this climate for this many years.
Collaborating with consultants and contractors
The system scales to teams because it is built on the folder structure and metadata, not on a proprietary archive.
When you pass the design to a structural engineer, the precedent library comes with it. They open the workspace and immediately see your research. They can search for what you found useful—tag:cold-climate-residential,load-bearing rating:>=4—and see the precedent buildings that shaped your envelope decisions without needing to import or re-document anything.
If the engineer finds a new precedent during their phase—a similar building's structural design, a case study of how others handled the snow load in your region—those go into the Precedents folder with clear naming. The watcher sees them land. Tag them, rate them, and integrate them into the conversation. Does this precedent suggest we need to adjust the wall thickness? Pull it onto the Design Research canvas and collaborate.
Contractors use similar reference. They care about construction sequence, material availability, and how this assembly was actually built on-site. A query like tag:construction-method,cold-climate rating:>=3 pulls every precedent about assembly strategies you have documented. That is what they need to estimate realistically. You do not need a separate "contractor reference library." You need one library tagged specifically enough that a query can slice into just the part of it another discipline needs.
Rating as you go, not in a second pass
This is the most important habit to establish early. Rating is not cleanup. It is the only thing that separates a searchable library from a folder.
The instinct is to leave rating for later—to get all your precedent photographs and details in first and rate them when you have time for a big review. But future-you has nine other projects on the board and will never get to that review. Instead, rate as you add a precedent photograph to the project folder, or five seconds after you photograph a material detail on a site visit.
Tag cold-climate precedents as you open their sections. Rating takes no longer than a quick assessment and it is the only thing that will make the library useful when you need it in six months. tag:cold-climate-residential images as you add them. Rating takes no longer than deciding "yes, this is a strong example" or "yes, this is instructive but has problems."
One month into your first project, you will have a library where the buildings you actually learned from are flagged as keepers, the ones you reviewed and rejected are marked down, and the images you never looked at are left at default. A search for any climate tag brings up the best versions first. That is the difference between a library and a folder you scroll through forever.
Building a library over decades of projects
Architecture is long. A building takes two years to design and one year to construct. Then it stands for fifty years and teaches you something new every time you pass it. Ten projects over a career means five hundred years of learning from buildings that exist.
Your precedent library grows because every project adds images. Buildings you visit. Details you photograph. Material samples you collect. Failures you document as you read about what went wrong elsewhere.
A library organized by tags becomes institutional knowledge. Search tag:seismic-detailing rating:>=4 when you start a project in an earthquake zone. The five projects that taught you about seismic design are there, with every detail rated and noted.
Search tag:facade-repair,before-and-after when a client asks why an expensive detail matters. Show them the building where a cheap detail failed after twenty years. That is the difference between a conversation and an informed decision.
The library is not organized by project. It is organized by learning. Each project adds more learning. Each search brings together every time you solved a similar problem.
How to organize references by project covers the folder structure that makes research efficient. How to rate and filter your best references digs into the rating system that makes searching faster over time. How to tag reference images consistently explains the tagging strategy that lets you slice the library a thousand ways. What is a reference manager answers whether you need a system at all.
Building a library of precedents is not different from building a reference library for any other creative work. The same principles apply whether you are documenting precedent buildings for residential design or collecting material studies for a commercial project.
The difference between a folder and a library
You already have a folder full of building photographs. A library is what happens when you stop organizing that folder by project and start tagging it by learning. The same photograph—a Vermont farmhouse with a standing-seam roof—appears in a query for "cold climate," another for "metal roofing," another for "seismic," and another for "additions to existing structures." It is one image that you photographed once. It earns its place in five different design conversations because you tagged it for what it teaches, not for the project it belonged to.
The reference app keeps it searchable. The tags make it work for you. And the ratings ensure that when you search for proven examples in a critical climate, you see the buildings that stood the test, not just the ones you photographed.
See how to build a reference library from existing folders to start with photographs you have already collected. The reference manager for architects describes how this specific workflow applies to your work.
Download the free trial and test this workflow on a real project. Thirty days, every feature, no card. The tagging habit is the foundation that turns a folder of pictures into a resource you can still use in fifteen years.