Paul van Oordt  SQL Services

Freelance SQL Server specialist and troubleshooter, Utrecht / Antwerp

+31 627400408 - WhatsApp -
LinkedIn - curriculum vitae - Nederlands

Recommendations for SQL Server development

Below is a checklist for the development of a SQL Server application. This list can be used when you develop software yourself, or when assessing that of others. The rules below allow you to create a maintainable and well performing application with an optimally enforced quality of data.

As to my services, I am not available for extended development work, but I can train a development team or assist it in modelling and writing the algorithms and code. Experience learns that the sooner I am involved, the higher the gain and the less work that has to be redone. A fresh outside perspective may benefit a system a lot. This is especially the case when adapting existing software. Some ideas on how the model could be improved may be around for years, but they might just not be the best choice.

Typically a project for me is one or a couple of days. Knowledge transfer is always an important part of the work. See also the possibility of an in-house SQL Server crash course.

Planning and design

When you are planning to build a SQL Server application, be sure that you are well aware of the possibilities that the product offers. You have to decide on the features below in an early stage, as including them later will usually require big changes. When evaluating third party software this gives you an idea of the level of expertise of the supplier, and to what extent the application is updated to use new SQL Server features.

This is an incomplete list of features that you need to know and understand to some extent, or that you want the third party to have used where relevant.

Modelling and database design

Note: The rules below also apply when you are developing code first using an ORM like Entity Framework.

When modelling and creating databases and tables the following is recommended.

Programming constructions

The following is recommended to create clear, correct, maintainable and well performing T-SQL code.


Most of the recommendations above not only benefit data quality and correct and maintainable code, but are also beneficial for performance. Neglecting these principles because of performance is usually a bad idea. Some extra recommendations to write well performing code:


(c) Paul van Oordt, Last modified: 28 December 2022