Assessment:Costs:Toolkit

This site is currently under development by the Cost Assessment Working Group. Please check back at the end of 2024.

In Sourcing vs Outsourcing

The case studies include information about what the institution is like, the collection being digitized, and the factors that affect the digitization method chosen - number of items, fragility, staffing, equipment, money, etc. The case studies explore what the initial choice was - insource, outsource, both - how digitization went, and a brief evaluation about the lessons learned about that approach. They discuss whether the project was entirely successful, proved too difficult as scoped originally, or had mixed results - some parts were completed well, but other parts of the project proved too difficult to complete. All situations are informative.

Case studies are available for review by original format and by approach/solution.

Original Format

Audio / Video / Moving Image

• Clemson University - AV Digitization
• Hoover case study - Audio tapes
• MIT Libraries case study - AVMI materials
• UMichigan - Super 8 film

Still Image

• Hoover case study - Posters
• UMichigan - Medieval Manuscripts and Scrolls
• UMichigan - Scrapbooks and Photo Albums
• University of North Texas - Yearbook Digitization

Text

• Hoover case study - Text-Based Archival Collection
• New York University case study - Access level (books)
• UCalgary case study - Administrative documents archival collection

Approach/Solution

Insourced

• Hoover case study - Text-Based Archival Collection
• UMichigan - Medieval Manuscripts and Scrolls
• UMichigan - Scrapbooks and Photo Albums

Outsourced

• Clemson University - AV Digitization
• Hoover Case study - Audio tapes
• MIT Libraries case study - AVMI materials
• University of North Texas - Yearbook Digitization

Hybrid/Blended

• Hoover Case study - Posters
• UCalgary Case study - Administrative documents archival collection

Undecided

• New York University case study - Access level (Books)
• UMichigan - Super 8 film

Institutional Information

Respondents were given the option to provide information about their institutions in relation to their case studies. The aim was to present this information to readers so they can see if their peers have similar circumstances and how they handled them.

Categories included: institution size, digitization experience, concerns about insourcing vs outsourcing, risk tolerance for outsourced digitization, risk tolerance for copyright, funding, IT resources, and skills available.

Hoover Institution New York University UMichigan

Scans per Linear Feet

The Cost Assessment Scans Per Linear Feet Subgroup was created to fact check current estimations published online for calculating images per linear feet. We also evaluated how differing digitization and housing practices as well as content type changed the amount of scans per linear foot. The results from this work can be used when determining the cost of a project using the cost calculator and/or other means of assessing costs like conversations with vendors.

Data

Further information about the survey, gathered data, and calculations can be found here: https://osf.io/ag9z3/

Content Type (Min / Average / Max) Linear Feet

• Photographs: 63 entries (658 / 1,871 / 2,515)
• Unbound pages: 120 entries (393 / 1,914 / 6,849)
• Mixed: Unbound pages, bound/folded paper documents: 15 entries (1,482 / 2,125 / 4,274)

Content Type (Linear Feet)	Number of Entries	Minimum	Average	Maximum
Photographs	63	658	1,871	2,515
Unbound Pages	120	393	1,914	6,849
Mixed: Unbound Pages, Bound/Folded Documents	15	1,482	2,125	4,274

Box Type (Min / Average / Max) Linear Feet

• Full Hollinger: 168 entries (103 / 2108 / 6,849)
• Paige/Banker/Records (filed legal): 64 entries (658 / 1,994 / 4,141)
• Paige/Banker/Records (filed letter): 17 entries (1,792 / 2,469 / 3,742)

Box Type (Linear Feet)	Number of Entries	Minimum	Average	Maximum
Full Hollinger	168	103	2,108	6,849
Paige/Banker/Records (filed legal)	64	658	1,994	4,141
Paige/Banker/Records (filed letter)	17	1,792	2,469	3,742

Images per Box Type (Not in linear feet)

• Full Hollinger: 168 entries (45 / 923 / 3,000)
• Paige/Banker/Records (filed legal): 64 entries (672 / 2,007 / 3,451)
• Paige/Banker/Records (filed letter): 17 entries (2,009 / 2,951 / 4677)

Box Type (Not in Linear Feet)	Number of Entries	Minimum	Average	Maximum
Full Hollinger	168	45	923	3,000
Paige/Banker/Records (filed legal)	64	672	2,007	3,451
Paige/Banker/Records (filed letter)	17	2,009	2,951	4,677

How to Use This Data:

General Use: The above data can be used to calculate and estimate the number of images likely to be produced by a digitization project. You can tailor your calculations based on box type, linear feet, and/or content type. Depending on your institution's processing practices, you may fall to one side of the min/average/max spectrum. Keep that in mind as you choose the number to use in your calculations.

Using with the Cost Calculator: If using the cost calculator on DLF’s website, the calculator asks for a number of images. Use the appropriate chart to estimate the number of images for your project. Just a reminder, the cost calculator is no longer being maintained so use at your own risk.

Using with the Cost Calculator Worksheet: The worksheet uses items and scans per item to calculate your number of estimated images. You can convert our data to fit this equation in multiple ways. Here are two examples:

Example 1: Column B represents the total amount of images you estimate using whichever chart is relevant to the data you have (boxes or linear feet). Since we’re calculating the number of images instead of items you would have a 1 to 1 ratio of images to “item”. So, column C is set to 1.

You have a 10 linear feet collection of Mixed: Unbound Pages, Bound/Folded Documents. You notice that your boxes are packed tight. In this scenario, it would be appropriate to select the max number on the scale, which is 4,274 images/linear foot. This means you would have 42,740 images. 42,740 would be the number for column B and 1 in column C.

Example 2: Column B represents the number of boxes you have. Column C represents the number of images you have selected from the chart Images per Box Type (Not in linear feet).

[Insert screenshot of spreadsheet]

You have a 10 Hollinger box collection of mixed: unbound pages, bound/folded documents. You notice that the boxes aren’t loose but aren’t packed tight. In this scenario, it would be appropriate to select the average number on the scale which is 2,951. In column B you would input 10, as the boxes will act as items, and in column C, 2,951 which is the number of images per item. Your total estimated image count would be 29,510.

Vendor Registry

The Vendor Registry group was created to get feedback from the library digitization community about their experiences with equipment and service digitization vendors within the cultural heritage sector. In 2023 the Vendor Registry subgroup circulated two surveys. Both surveys provided a list of vendors curated from a previous vendor registry, as well as provided an opportunity for respondents to add vendors to the list. Respondents were also asked to indicate whether they would be available to be contacted directly to answer questions.

The subgroup compiled the survey results into an overview document that listed each vendor, its website, a description of product or services received, and overall service experience. These results were then made publicly available via Reformatting Service & Equipment Vendor Registry Survey Results. The Survey Results document also included an indicator whether there were survey respondents that would be willing to share their contact information with others to provide additional details about their experiences.