Each time a request is sent to the server, we count it as 1 operation. It includes index, autocomplete keystroke, search and etc. For example, if user is attempting to type in “travel agency” to the search bar, per each character user typed there will be a request to our server for suggestion. Often time, user doesn’t need to finish typing the full phrase before they can spot the result they want. For our data, it indicate it takes in avg 5 characters for a user to locate what they are looking for. That is to say, if you divide operation counts by 5, you can roughly obtain the number of search intents from your users.
For paid users, we will pick the main image per post in your website, resize, optimize and store them in one of our image servers. Images on our image servers will be distributed in CDN network that will improve the serving speed for our autocomplete and result page.
- For non-paid WordPress websites, we will create thumbnails in their server and images will be directly served from their own server instead.
- For non-paid regular websites, we will not create any thumbnail but resize their original image photo on the fly during serving. If the original image size is large, users may notice that.
Full term match is one of the matching option supported by SearchIQ. It lets a user query matches your posts/products whenever “all terms” in the user query is found on the posts/products regardless the ordering. This option allows you to increase the coverage of your posts/products but still keep a high relevancy.
Broad match is the default matching option that SearchIQ uses. It lets a user query matches your posts/products whenever someone searches for that phrase, similar phrases, singular or plural forms, misspellings, synonyms, stemmings (such as floor and flooring), related searches, and other relevant variations. This option allows you to increase the coverage of your posts against user query but it also gives a chance for some not highly relevant posts and products to be shown. To migrate the irrelevancy issue, SearchIQ will rank the most relevant posts and products at the top of the result. Some of our users use this option along with field weight specified. This gives them better control the ranking of the matched posts/products.
Phrase match is one of the keyword matching options SearchIQ provided. It allows you to match user query in exact form from any of the fields indexed. For example, if user types “tennis shoes”, the posts/products contain “tennis shoes” with other terms before or after the phrase will be shown, but not for “shoes for tennis”, or “tennis sneakers”. Phrase match is more targeted than broad match and full term match.
However, if you feel this match option is too restricted. You can consider to use broad match and full term match. Both of these matching options support phrase match with quotes before and after the search query.