Starting in earnest in the late 1990s, we began to see an increasing number of online digital archives containing digital versions of historical primary sources. The “digital surrogates” as we call them, may have been in the form of digital photographs of the original, or perhaps transcriptions of the text stored in a searchable database. These websites make primary sources accessible to anyone, no matter where in the world, and the digital format allows new types of research – linking, network analysis, corpus linguistics – like never before. Many contain huge collections that run into the hundreds of millions of words. They are archives in their own right. We have come to love the beautiful and user-friendly search interfaces that point us to relevant sources. We admire (even envy) those colleagues who win the big grants to put these sites together. We have seen all the positives, but in the process have forgotten to ask a fundamental question: are these websites necessary?

Not, are these digital records necessary? I believe they are; or at least, that they provide tremendous benefits that justify their creation and their cost. Instead, I argue that we must question whether or not a website is always the best (most economical, sustainable, good use of funds) way to store and distribute these records. Sometimes the answer is yes. Often, depositing your digital records in an existing repository, or releasing them in a standards-compliant format for download, provides all of the same research benefits as would a website, but without the hassle and expense. Every time a website is built, less energy and money is available to digitise more materials. That’s not good for taxpayers, and it’s not good for researchers. Instead, I argue we should be critical of the need for a website, and should spend more time making data available, and less time making it look pretty.

Of course, there are plenty of good reasons for building a website for your collection, including learning a new skill, protecting fragile resources from constant handling in the archives, adding interactive functionality that is only possible on the web, and opening access to users who cannot visit in person. But often there are better ways to share your material. Websites are expensive and a lot of work. Committing to building a website is like committing to build and maintain a library for the foreseeable future. If you are reading this, you must already be enthusiastic and have a great idea. This flowchart is not meant to dampen that enthusiasm. Instead, it was created to make sure you ask yourself some of the tough questions too, to make sure your project is viable and necessary before you make a big commitment to transform your collection into a database-driven website.

Before taking on a major web-based project, I urge would-be website builders to pass three tests: morality, audience, and competency. These tests are outlined in greater detail in the flowchart that accompanies this editorial. The morality test asks the reader to consider the cost of the project and the value for money of a web-based solution. The audience test is a series of questions designed to tease out which users the website seeks to attract (journalists, family historians, students, academic historians, the general public), and by extension of that, if a website is the best way to reach that audience. Finally, the competency test ensures that you or your team have the technical experience to build an effective, sustainable website that can compete for the attention of users for many years to come.

Large database-driven websites containing historical materials are a great asset to our society and to academia. All I suggest is that we make sure that the ones we build are worthy of the time and money they take to produce.