The new highlight search is a great new feature. However in its current implementation the performance is poor compared to a search dialog search on an exact string match.

For example.

A string search on our database for a customer reference in a regular search dialog takes around 0.3 seconds. A search using the new highlight search took 20 seconds!

I believe the issue is that the search type used in the new highlight search feature is using a LIKE query which in turn is also passed to the SQL database as a LIKE query also, making the database search inefficient. We have good indexes on our SQL database but over 10m document rows. Staff wont use this feature with that level of performance.

See the URI in the browser being used below which hints its a %LIKE% search;

https://REMOVED/DocuWare/Platform/WebClient/Client/Result?rl=REMOVED&fc=REMOVED&q=%2520%5BPOLICY_REFERENCE%5D%2520LIKE%2520%2522%2ABUNX22VA01%2A%2522%2520&queryInInvariantCulture=True&displayOneDoc=False

A better implementation would be to allow the configuration of the search type in the highlight search configuration such as = / CONTAINS etc.