I've had this nagging feeling for quite some time that much web development time is wasted in designing and developing search boxes that are never used beyond the developers and testers.
With browsers implementing search bars themselves, I rarely use web-page level search boxes, with the exceptions of search engines and online stores. For blogs and informational websites, I often don't trust the built-in search to return better results than I'd get from Google.
No matter how perfect your site hierarchy and navigation is, some users won't understand it. Or they don't want to learn it.
For those users, a search box is paramount, because that's the primary way they use the web. To them search is navigation.
Rather than exclude those users from your site, leave in the search box. It doesn't hurt the usability or appeal of the site for people that don't use it.
site:ux.stackexchange.com searchterm
Commented
Mar 3, 2012 at 10:47
site: keyword. For them having to go back and google something might seem like a lot of work with no promised solution
In short: Depends on the context. If the site belongs to a business: Very important.
I'm not aware of evidence suggesting the your premise that search boxes are only used by developers and testers is true.
I would argue that there is little point of having a generic Google search box displayed on your page, unless your business model relies on your site being used as a portal. Many internet providers still do this on their pages and suggest this as a default home pages for their users.
In-browser search (ie ctrl+f) will just search for strings on the page you're looking at, so that's not sufficient usually. Leaving the page for a search engine doesn't limit the scope to the site you looked at last unless you use site:example.com type syntax to limit your search. Few of the users you want to help with the search box would know how to use this though, so it's not helpful if the user wants to find the information on that site, rather than on any site.
From a business perspective you don't want people leaving your site to do an external search! It's quite possible that your competitors may appear higher in the list of results. Your search function should ultimately bring users closer to a sale, or the information they're looking for, without them having to leave the site. This is just as true for information sites that may make money with advertising (pageviews!) as with eCommerce sites that make money with products.
ctrl+f) will just search for strings on the page" what relevance does this have to the question? I was referencing the search box (sometimes built into the location bar) in the browser. It's also visible on the mockup I posted.
A bit of data, albeit a little old, from the venerable Jakob Nielsen:
Our usability studies show that more than half of all users are search-dominant, about a fifth of the users are link-dominant, and the rest exhibit mixed behavior. (Search Usability, July 1997)
Unless that number has changed dramatically over the last decade, a good number of users are potential searchers. Omitting a search feature entirely could sorely disappoint such users.
Update (thanks, zzzzBov): The age of that data is troubling, but in an Alertbox from 2005, Nielsen makes an interesting point (I can't believe I'm quoting Nielsen twice in one day):
Earlier guidelines for search usability continue to hold, and are becoming even more important with the new mental model. The dominant search engines comply with all the main usability guidelines, which is obviously a major reason that they're on top. Today, the guidelines don't just describe good search; they describe expected search. (Mental Models For Search Are Getting Firmer, May 2005)
In other words, because of the rise of Google et al., users expect a search box whenever they need one that lets them type in any term and magically get to just what they're looking for. Omitting the search box, then, runs the risk of violating users expectations.
The search box is often one of the most overlooked, and neglected parts of the user experience.
From a data mining perspective, every time someone types a single thing into that search box, they are giving you insight into how they see and think about your site.
From a sales/business side, mapping those searches to your site content, and presenting unique experiences is what can be the difference from a good or a GREAT site.
Examples:
Amazon, predictive input reduces your likelihood of every performing a search with no results.
Zappos, search on a brand, and you get taken to a customized results page solely focused on that brand.
The easiest way I think of it is make sure your search experience supports wide and narrow searches. Tune your content accordingly (blog - make sure articles are found, appear high in results) (e-commerce - products map to the broad cross-section of keywords).
Monitor your analytics, and watch conversion/time on site stats jump!
Enjoy!
This research by "think with Google" titled "Principles of Mobile App Design: Engage Users and Drive Conversions" from March 2016, gives a series of principles for apps usability based on studies of "more than 100 people on 100 different apps".
We partnered with AnswerLab to conduct a user study of more than 100 people on 100 different apps across a variety of verticals including e-commerce, insurance, travel, food ordering, ticket sales and services, and financial management. (Gaming apps, social networking apps and music services were not included in the study.)
The study is focused on Mobile apps.
As many as 25% of app users open an app once and never return (Source: eMarketer “App Marketing 2015: Fighting for Downloads and Attention in a Crowded Market,” July 2015).
More than ever, people are engaging with their phones in crucial moments and for shorter periods of time. Their experiences need to be efficient and delightful. Plus, a well-designed app that provides utility has the power to cut through the clutter. It can deliver on people’s many I-want-to-know, I-want-to-go, I-want-to-buy, and I-want-to-do moments.
In Chapter 2 it gives several principles about the use of search functionality in Mobile apps.
In-App Search
Effective and useful app search is crucial for helping app users find what they need. (...)
This point is significant about the search box:
6. Prominently display the search field.
Users with a specific task or need will typically look for a search field. They often prefer this to browsing. Apps that do not have a prominently placed search box can cause user frustration and slow the user down.
Don't:
The search functionality is hidden behind a menu option.
Do:
An exposed search field is easily located.
I only included the point relevant to the search box. Other points in the document include principles about search results, search filtering, etc.
Additionally, if you remove the site search box, you're sacrificing FOR YOURSELF a key source of knowledge of what your users want from the site: the ability to analyze their on-site searches.
No place else on your site can you understand in words exactly what they think they'll find someplace in there.
Highly recommend Lou Rosenfeld's "Site Search Analytics" text.
A search option should be provided on all pages where it may be useful -- users should not have to return to the homepage to conduct a search. Search engines can be helpful on content-rich websites, but do not add value on other types of sites.
The search box is not a substitute for good content organization.
According to a study conducted by the U.S. Department of Health and Human Services this is in the top tier of importance on Websites. There is also much supporting evidence from other studies, including Farkas, Nielsen and Levine.
Farkas, D.K Guidelines for designing web navigation - Technical Communication #47
Levine - Sun Microsystems guide to web style
Nielsen - Ten good deeds in Web design http://www.useit.com/alertbox/990530.html
U.S. Dept of Health and Human Services - Research-Based Web Design & Usability Guidelines
It depends how big your site is - the bigger the site (and the more 'stuff' which is on the site) the more important the search box is.
ie - imagine using Amazon - without it having a search box...
Our analytics show that a bout 2.26% of users use the site search. Of course we push for google search and SEO and most people landing on the site is doing so direct to their search result. I personally hardly use search boxes and with responsive design it creates extra work and rethinking of header.
It is really depending of the content and the size of the site. Also if you have many documents, attachments, reports in your web site considering that they have been tagged properly site search is a must. Or if you have many products in your e-commerce portal... I don't think I can try to figure out clicking to find cookies(real,yummy ones) inside Amazon. (why would I do that?)
And even in our case, yes we should not omit it. One of the side site I designed I did not use it because it is clear to go through the main and home page navigation. Idea was to segment users from the start. Users can not search if they are not sure of what the site is about.
In 2001, Jakob Nielson published an article that explains the usefulness of search box and he summarized " Search is the user's lifeline for mastering complex websites. The best designs offer a simple search box on the home page and play down advanced search and scoping".
Users love search for two reasons:
Search lets users control their own destiny and assert independence from websites' attempt to direct how they use the Web. Testing situations routinely validate this. A typical comment is: "I don't want to have to navigate this site the way they want me to. I just want to find the thing I'm looking for." This is why many users go straight to the home page search function.
Search is also users' escape hatch when they are stuck in navigation. When they can't find a reasonable place to go next, they often turn to the site's search function. This is why you should make search available from every page on the site; you cannot predict where users will be when they decide they are lost.
Search is a big deal and based on the research conducted by Jakob Nielson and Norman groups, titled " Intranet Usability: The Trillion-Dollar Question" and conluced that poor search was the greatest single cause of reduced usability across intranets we have seen, aside from the general lack of executive support and budget. Search usability accounted for an estimated 43% of the difference in employee productivity between intranets with high and low usability.
In an attempt to prove the usefulness of a search box in the website, He"Nielson" replaced a link in one of the website by a search box and that increased by 91% of the usage.
Search On Mobile Devices ( Web and Native app).
To avoid opinionated answers, I am only going to list down number of research about search, the magnifying icon, the autosuggestion and other 62 design recommendation for both web and mobile devices.
All of these research are dated on 2015 and done by the top ux experts such as Nielson and Norman.
References :
I suspect I'll be repeating some of the stuff but here are my views.
The questions you will need to ask yourself are :
1) How big is the site : If the site is really big and you have a lot of content which has to be searched then you surely need a search box ,however you might also need a search box in cases where your site might be small (say 4-5 pages) but might contain content like specific reports or might be a recipe site about Indian food. So if a visitor comes to your site and he is looking for a specific dish but he is not sure what category it falls under ,he is more likely to use the search
2) How is your information architecture/site navigation defined : If your site architecture is such that users need to have a lot of clicks to get down to their content ,then you should consider a search. You should also consider a search if people are unlikely to dig up content unless they specifically know where it is or it buried fairly deep or if there are a lot of frustrated clicks
3) How much time are users spending on your site pages : Now I admit ,I cant get this information without doing some site analytics but if the average user spends a short amount of time on your site,you will need to consider if they can actually assimilate the content on the site and find out where to go to get to the content they are looking for.Of course ,if you have a very engaging site ,you might not have this concern but with low attention spans prevalent nowadays, how would you be sure that users are not going to give up and navigate away from your page
4) To quote what J. Jeffryes said, users are accustomed to navigating around sites using a search box because thats the mental model they have established with regards to information retrieval. Breaking that mental model and asking them to conform to your navigation hierarchy might frustrate them.
5) Though you did state that your trust level of the search engines of blogs and informational sites was relatively low, you are (hopefully) not the only user of the site you are designing and there might be other people who would have different views about the efficiency of the custom search engine is. Also a lot of companies nowadays use google search built into their sites and the search experience might turn out to be really good
6) Lastly echoing what Mike Hill said in one of the answers, if you have incorporate predictive search,you would help users get an idea about the content available also reduce the amount of directionless searching.
Lastly from an anecdotal point of view, I work with a small design team in my company and the client has a requirement that any sites designed for them should use content only from their media site. Now the search for the media site is really bad and incredibly unusable but everyone in the team would rather prefer to use the search to find the content despite the irrelevant stuff that keeps coming up despite it sometimes seeming easier to just drill down and find the content via the navigation
It is purely depends on the requirement. The search box implementation to the website is a additional feature to make user more comfortable with your website, if you have more pages and contents.
In this case we should have search box to search the content instead of rolling the contents using mouse or keyboard.
If the website is a E-Commerce / Blog / Media related, we defenitely implement search feature there. The main reason to implement this feature is usability and depends on the website we going to work.
Thanks.
It doesn't seem like an extremely important thing when you just have it by default, however it might turn out to be extremely important if you remove it.
You'll know if this is something crucial for your site/app very quickly if you remove it.
Another important thing to consider is not to go crazy about custom search forcing users to for example choose what exactly to search for (e.g. articles, videos, etc.) as this is not particularly useful, rather confusing. "Don't make me think" by Krug makes a good case about it.
The search box can prove to be very important on your site, not only for your users to find content they are looking for, but also for you to see what your users are looking for...which may indicate a content problem on your site. If users are consistently looking for similar keywords, this may be a good opportunity for you to take at look at how your site is organized and even how your copy is written.
If your site search from is powered by Google Custom Search you can track analytics on keywords users have searched for within your site. Yes, some users will take the lazy way out and automatically default to using the search box to find content, but it's good user experience to give them that option rather than expect them to navigate the labyrinth of large sites.
But if you see that your users are relying on your search form, you may need to revisit your site structure from a user experience point of view.
As many have said already, it really depends on the type of the website.
Content heavy websites for e-commerce, collecting documents, asset management, real estate, heavily rely on search to allow users to find what they are looking for. There is a lot of research out there, books, blogs that cover this area.
Marty Hearst : Search User Interfaces
Designing Search: UX Strategies for eCommerce Success
Sites with a simpler nagivation hierarchy and less content to search for usually don't invest in site level search and I believe it is for a good business reason. In these cases, search is more of a "nice to have" feature that sldom makes it to production.
This kind of decissions have also lowered the confidence of the users that search will give them any good results, or any results at all.
For the question in the bounty: "I am curious whether the answer to this question has changed with the marked increase of mobile website use in that period."
According to Material design search should be present when an app supports large amounts of information. So the main idea is that the search action should be present where there is a significant amount of content. Depending on the way the app the search can get more or less importance in the UI.
When an app supports large amounts of information, users should be able to quickly locate content by searching for it.
(...)
Basic search involves:
Opening a search text field
Entering and submitting a query
Displaying a set of search results
There are two types of search patterns depending on the importance of the search action:
Persistent search
Use persistent search when search is the primary focus of your app.
For example Google Maps where the search action is the main action.
Expandable search
Use expandable search when search is not the primary focus of your app.
For example Google Keep where the user can navigate through the app and use the search action if needed.
Material design guidelines was developed by Google in 2014 and is still being updated with new content.
There is an exelent article with sources of different researches. I'll extract the elementary points in the question but I recommend to check out the whole article: Search vs. navigate: How do people behave on websites?
So to quickly answer your question, this is the important point:
So, do you really need internal search box on your site since we “already have” Google now? If your website has an advanced structure and significant amounts of content, then the short answer is YES, because in order for any website to be effective, it should be easy for visitors to engage with it and find the content they are looking for — and people are different and use different methods to accomplish tasks.
Do you need a search bar on your own website?
Think it differently: When a user visit your site, does he search something specific (e.g a product) or does he only explore your site or want to know your prices (e.g if you're a agency)?
If the answer is specific, then you should implement a search-function to ensure, your user reach his goal fast.
And there are user who are more likley to use the navigation, other use rather a search bar. So don't exclude the second type of user on a complex website.
You want some numbers to prove that users are using an inbuild search? Here's another quote from the article:
According to Comprend 59% of web visitors frequently use the internal search engine to navigate on a website (...)
If you want to know more, make sure you check out the linked above article. He covers everything you need to know. Oh, and it was written in 05.2016. So it's newly researched.
Another great article about why you should provide an enhanced on-site search: http://bma-milwaukee.org/press-releases/the-value-of-enhanced-on-site-search/
And a great article, that a better on-site navigation is more important: https://www.nngroup.com/articles/search-navigation/
Developers design and implement search features because it is an easy thing for them to do. However, often there is no rhyme or reason for implementing search.
Peter Morville & Jeffery Callender wrote a great book called Search Patterns: Design for Discovery, and this shows there are many search and findability patterns, each serving specific situations, goals and motivations. There are also some anti-patterns you need to be aware of.
Ideally you will perform some user research where you speak to real people, to help you discover their specific needs. The results of this user research may surprise you.
They have a website to accompany the book: http://searchpatterns.org/
Search is among the most disruptive innovations of our time. It influences what we buy and where we go. It shapes how we learn and what we believe. This provocative and inspiring book explores design patterns that apply across the categories of web, e-commerce, enterprise, desktop, mobile, social, and real time search and discovery. Using colorful illustrations and examples, the authors bring modern information retrieval to life, covering such diverse topics as relevance ranking, faceted navigation, multi-touch, and mixed reality. Search Patterns challenges us to invent the future of discovery while serving as a practical guide to help us make search applications better today.
This is kind of an open ended question which is very difficult to answer comprehensively without information about what it is you're trying to do. As has been noted in numerous answers, I think you're unlikely to find gospel truth on this topic. Research that is true for some websites (eBay, Amazon, etc.) may not be true for others (WordPress blogs, for example), and your target demographics and implementation complicate this further (hypothetically, an eBay clone targeted at geriatrics may not see as much searching and more reliance on menus, and if your search is mediocre and your menus are amazing and intuitive, people may gravitate to those naturally, skewing your metrics). I'll post some trade-offs and considerations since other answers have some decent examples on research.
The obvious are thus:
+ Your site is simpler in design and has less visual clutter.
+ There is less work required to implement your website.
- Users who like to search can't search anymore.
The importance of this varies depending on the site. Generally speaking, considerations could include:
And so on.
As with pretty much all these types of questions the answer has to include "it depends"
Personally I design as though the site search doesn't exist, but very rarely would it be something I'd recommend removing.
Having facilitated hundreds of user testing sessions, when given a task there is a recurring answer from users which is "I would just use the search". Testing is a bit of unrealistic scenario, so it may well be in a lot of those instances they would actually have performed the search external of the site using Google/Bing.
I have also identified that there are typically two different types of participants: "searchers" and "navigators". "Navigators" will always try to navigate via the menu structure and hyper links presented to them. "Searchers" are more likely to gravitate towards the search bar, with an expectation that it will solve their problem (Personally I expect them not to work particularly well!). Obviously there are mixed behaviours dependent on the task.
I would think very carefully before considering removing existing search functionality. Do some testing and look at the metrics to understand whether it is being used.
Further reading:
https://www.smashingmagazine.com/2008/12/designing-the-holy-search-box-examples-and-best-practices/
http://boxesandarrows.com/search-behavior-patterns/
https://www.nngroup.com/articles/search-and-you-may-find/
https://www.cludo.com/en/blog/search-vs-navigate-people-behave-websites-search-navigate/
Since the question was asked more than 4 years ago with a lot of answers proposed but none was accepted, I assume the author is expecting something else. A survey of existing answers generally fall into two groups:
Regardless of the choice, conclusion is often made by comparing with another UX behavior - navigation. Both behaviors assume user has already landed on the site. The question, however, seems suggest that comparison should be made against external search services such as Google. Indeed, the question that ought to be investigated is: with increasingly improved accuracy of external search engine, what percentage of successful search is achieved within the app as opposed to user hitting the destination page referred to directly by external search engine after entering a query of which specificality is progressively improved with the assistance from the external search engine, taking into account information that is unavailable to in-app search such as user's location, historic overall search and navigation behaviors etc?
Unfortunately I haven't found any relevant up-to-date academic-grade research. But from both personal experience and from the fact that some popular sites such as the following have no site-wide in-app search feature, yet no usability nor SEO ranking is affected
I tend to believe for CMS and blog type of web sites of which core content is not too volatile and accessible by general public anonymously, hence crawlable by external search engines, site-wide in-app search is becoming non-essential.
My own research ( as yet unpublished ) indicates that the search is the MOST important part of the site, even more than the navigation hierarchy, which most sites give a whole lot more time to. When users have come to a site home page, the most common route to progressing is the search.
Of course the alternative is that they come via another search engine to a specific page. So it might depend on where users drop onto the site from as to how important it is. I would surmise that for repeat customers - who are the lifeblood of an ongoing business - they would often start with the home page and progress, rather than new customers who may come via search engines looking for specific products.
The answer depends entirely on your site and your site's users.
You can't know the answer unless you've tracked your user's behaviors on your site. Tracking tools such as Hotjar offer you insights into what your users are doing on your site. It offers you the ability to create funnels to allow you to see how your users navigates from page to page and where they are having issue.
Other analytics tools such as Optimizely allow you to test out no search vs. search or prominent search vs not prominent search box on your site to determine which method works best for you.
There's no quick answer to your question unless you do the research and find out for yourself if search is relevant on your site to your users.
