What S1000D
really costs.

What does S1000D cost?

An S1000D project costs in three places: the software, the work of turning existing manuals into data modules, and the upkeep once the modules are live. The software is the part with a price list. The work is the part that decides the budget, and nobody publishes a price for it.

The specification itself costs nothing. It can be downloaded free of charge from s1000d.org. Everything around it is where the money goes.

What does the software cost?

A project needs a common source database (CSDB) to keep the modules, an editor to write them, and a publishing engine to turn them into manuals or an interactive electronic publication. Most vendors quote these on request. One has published list prices: Contiem, on the UK government’s G-Cloud 15 catalogue, for its notusCSDB.

These are one vendor’s catalogue prices, before setup, migration and training. They are a useful anchor: for a team of five to ten writers, the software from this vendor runs to tens of thousands of pounds a year, not millions. Which tools exist for each job, and what to ask their vendors, is in how to choose S1000D software.

What does the conversion cost?

There is no credible public price per page for converting manuals to S1000D. Service providers quote per project, and the percentages of savings found on vendor sites come without a method. Anyone who names a price per page is quoting their own offer.

What can be said is where the effort sits. The S1000D consultancy ONEIL sums it up: “conversion isn’t just mapping, it’s restructuring”. Every page of a supplier manual has to be analysed, split by type of information, rewritten into the structure of the specification, coded, illustrated and checked, as described in how to convert legacy manuals to S1000D. These are the drivers that make one project cost twice as much as another of the same size:

For ships, the volume is the first number to know. By our estimate, the documentation of one ship type rests on around 200 supplier manuals, keeps about five writers busy full-time and costs in the order of €2 million.

Does S1000D pay off?

The best public evidence is a cost-benefit study by the Institute for Defense Analyses for the US Department of Defense, published by the Naval Postgraduate School in 2010. It looked at moving the Navy’s technical manuals and training material for hull, mechanical and electrical equipment, more than 45,000 pages a year, onto S1000D and one shared repository. Against an investment of $8.7 million, it put the net benefit at $78 million over ten years in its base case, within a range of $32 million to $160 million.

The benefits come from two places. Content is written once and reused across manuals and training. And changes are made once and reach every publication that uses them. The scale this reaches is visible at NAVSEA, whose enterprise S1000D solution held more than 1.5 million pages and 70,000 illustrations by May 2018, according to the Navy’s own report.

What does it cost after go-live?

Documentation is never finished. Suppliers revise their manuals, equipment is replaced, and every change has to reach every module it touches. On a ship that lives for decades, the upkeep can outweigh the first conversion.

The cost of an update is mostly the cost of finding what it affects. If you know which module rests on which page of which supplier manual, a revision becomes a list of modules to check. If you do not, it becomes a search.

How to estimate your own project

The only reliable numbers are your own. Convert one system end to end, with every type of module you will need, and measure. Then scale with these inputs:

1. Number of supplier manuals and total pages, split into scanned and digital.
2. Expected number of data modules per type, from your DMRL.
3. Hours per module of each type, measured in the pilot, including review.
4. Number of illustrations to redraw and parts figures to link.
5. Software and infrastructure, per year.
6. Revisions per year, and the hours to process one.

Hours per module times modules gives the conversion. Revisions times hours per revision gives the upkeep. The pilot also tests your business rules before hundreds of modules depend on them.

How to bring the cost down

• Agree the business rules and the BREX before the first module.
• Keep warnings, tools and supplies once, in common information repositories, and reference them.
• Ask suppliers for their manuals in digital form, not as scans.
• Let software do the reading, the first draft and the checks against the schema and the BREX.
• Keep a link from every step to its source, so that review and every later update start from the right page.

This is the part Tracium takes on. It reads the supplier manuals, drafts the modules, checks them against your rules and links every step to the page it came from, on your own servers. The people stay where they are needed: on the decisions and the approval.