Is your search experience leaving you a little unsatisfied?

Give these Search Tweaks a try. This site has eighteen tools for enhancing Google search in four categories -- Query Builders, News-Related Search, Time-Related Search, and Search Utilities. Some tools, like Back that Ask Up, make existing Google features easier to use. Others, like Marion's Monocle, add search functionality. Hold your mouse over each menu button to see a popup explainer of what a tool does. If you like what you see, give the button a click.

Using this site requires JavaScript. It's designed to work on desktop. It should work on your phone but the design does not anticipate that. This site uses Simple Analytics because privacy, it's a great idea.

None of these tools use the Google API. Nor do they use scraping. Where's the fun in that?

Search Tweaks is a Calishat jam! You can reach me at Mastodon or LinkedIn.

Wiki-Guided Google Search Takes Your Wikipedia Topic Search and Builds a Set of Google / Google News Queries Around Related Topics.

Say you want to search Georgia O'Keeffe but you want to start with something more specific than a general Google search for her name. Wikipedia knows a LOT about Georgia O'Keeffe, so why not use it to make a more specific query?

Start by entering a topic you want to search. You can search anything with a Wikipedia article. Wikipedia articles start with the article title and then the phrase "From Wikipedia, the free encyclopedia," like this: "Cooking From Wikipedia, the free encyclopedia." The part before the "from Wikipedia" phrase -- in this case, Cooking -- is the article title. You can use any Wikipedia article for your search, though the number of mentions each page has will vary by the page's popularity. They're both good rappers, but Queen Latifah will have more page mentions than Qveen Herby.

Then choose how many times your topic should be mentioned on a Wikipedia page before it's displayed in the results (from 1-20). I recommend starting with 2-3 mentions when you're not sure how popular the topic is, and 10 or higher when you know it's a popular topic or a common noun. (Beer, for example, is always going to give you a long list even if you restrict your results to pages which mention beer at least 20 times.)

Your search results will include a list of pages where your topic is mentioned with links to Google and Google News searches for each. The Google searches will include both your original topic and the related topic, giving you a list of focused searches around your original topic without any additional expertise or understanding required by you.





Clumpy Bounce Topic Search Uses Wikipedia Categories to Build Topical Google Searches

Imagine you pick a news article at random and start counting the number of times a baseball player is mentioned by name. If one player is mentioned, it might be a cultural reference or a quote out of context. If two baseball players are mentioned by name, it's possible that the context is quoted dialogue between the two of them. But if three baseball players are named, then it's getting likely that the article is in fact about baseball.

That's the premise upon which Clumpy Bounce is based: if multiple people, places, or things relevant to a topic are mentioned on a Web page, it's more likely that the page has something to do with that topic. Clumpy Bounce uses Wikipedia categories to make lists of popular pages within a category. It then builds Google queries based on the pages you pick from that list. Here's how it works, in three steps:

1. Enter a keyword search (anything that might be in Wikipedia) and you’ll get a list of categories associated with that keyword. Single keywords are best.

2. Pick a category and you’ll get a list of checkboxes showing you the most recently-popular pages in that category.

3. Pick up to three and Clumpy Bounce will “clump” them into a Google search along with some cruft-reducing anti-search and “bounce” you to a new tab of Google search results.

The idea is to direct to rich, information-dense results in Google.

Try searching for company names, or plants, or minerals, or even fictional characters. Often you'll see very famous people who end up at the top of categories with which you do not normally associate them. Don't add those to your searches or you won't get great results. If you don't know anybody or anything on the page list you generate, pick a couple in the middle.

Try a search!



Smushy Search Helps You Build Topical Searches on Google Without Getting the Same Old Results.

When your knowledge of a topic is minimal it's hard to refine a search query to get past the very general "front door" results that a search engine can provide. Enter Smushy Search.

Smushy Search takes a topical search term and uses the Datamuse API to find related words (a frequency setting allows you to specify how common the words should be). Four of the found words are then picked at random, added to your query term in two "or" pairs, smushed into a Google query, and opened in a new tab. The idea is to let you explore a topic with in-depth search results without needing any kind of prior knowledge, and without the time needed to build a specialized vocabulary.






This will remove Amazon and eBay, and attempt to remove common ecommerce results.



Does what it says on the tin. Might break your search, but worth a try.



The lower the frequency, the less the word is commonly used. Low-frequency words generate the most-focused results but there are fewer of them to find.



Find News Sources on Wikipedia and Search Them Via Google With Non-Sketchy News Search

When you search Google News, you're trusting that all the sources included in your results are legitimate. Often that's the case, but there are a lot of bad actors trying to poison Google's search results with disinformation, propaganda sites, and other such garbage.

Non-Sketchy News Search uses SPARQL to find media outlets on Wikipedia by keyword, then bundles them into a Google search. Presence in Wikipedia does not guarantee legitimacy, of course, but at least you know what sources are being searched.






After you run the search you'll get a list of media outlets that match your search terms. Click on the checkboxes of the outlets you'd like to include in a Google search (you're limited to 15) and click the button at the bottom of the list to open a Google search in a new tab.

Loading...
Select Source Description Website
Marion's Monocle v2 finds American TV stations by state (via the FCC Licensing & Databases Public Inspection file), groups them by city, and lets you search their Web space with Google as well as find recent published news stories via Google News. Start by using the pull-down menu to find a state.

No Stations Selected No Stations Selected
Choose a state from the drop-down menu and the city menu will repopulate with a selection of cities.

Choose a city and you'll get a table of schools in that city. Each row in the table will have a link to search that school's name on Google News, with and without topic modifers.

Please remember that Google News' results reflect only those stories which have been indexed, so please do further research if you find a place or a situation which sparks your interest.

But right now, start by selecting a state: City list:

Do you want local news? I mean LOCAL news?

Enter a US address. StreetScoop will find the nearest large city, query the FCC for the television stations in the area, and aggregate those domain names along with your street name into a Google search, which then opens in a new tab. If the address you entered is not found you'll get an error.

Please don't use abbreviations for street types -- use Street not St, Road not Rd, etc. Do not use suites, apartment numbers, or commas in your address. If your address is not found you'll get an error message. Sometimes it's finicky for a reason I haven't determined yet and I apologize.

Many thanks to SimpleMaps for the dataset that I needed to get this doing right(ish).
 


Back that Ask Up Quickly Removes Recent Content from Your Google News Searches

Enter a Google News query and the number of days/months/years' worth of the most recent news items you want to eliminate. This tool will construct a date-restricted search query in Google News and open it in a new window.

New political shenanigans disrupting your search? Back that ask up! A celebrity feud mucking up your results? Back that ask up! Just don't want to remember the last couple of days? BACK THAT ASK UP!

BTAU won't work perfectly because sometimes dynamic content breaks the date-based search. But it will eliminate a lot of recent news articles.

TimeCake Makes It Easy to Create a Set of Time-Bounded Google Searches

Here's how it works. Enter a starting year (minimum 1999) and ending year, along with a number of years to use as your interval.

TimeCake will generate a set of date-bounded Google searches based on your input. Click on one and it'll open in a new window.

Very useful to explore the development of a topic over time.











Obit Magnet takes a name and death date and generates date-bounded obituary search URLs.

URLs generate for Google Books (newspapers only), Newspapers.com, NewspaperArchive.com, and Chronicling America.

Enter the name of the person you're searching and the day, month, and year of their death. Obit Magnet will generate news searches for a span of 7 days and 15 days after death. The resources that can be specifically set to search for obituaries will be, while the others will have "obituary" added as a query word.

Names: If the person you're searching has a middle name, use it in the search. Obit Magnet will automatically create searches with and without a middle name. If you're searching for a married woman and you have her middle name, enter her full name like this: firstname middlename maidenname lastname. Obit Magnet will automatically generate a full complement of name variations for your search.

Sources: Neither Google Books nor Chronicling America require registration to access their results (though some Google Books results may be behind a paywall.) Newspapers.com is a pay site but is generous about letting you view search result pages without being logged in. NewspaperArchive.com is a pay site and if you're not logged in can sometimes generate weird search result pages. Try the other resources before NewspaperArchive.com. One more thing: Chronicling America has nothing after about 1963, so don't bother checking it for more recent obituaries.






Software VerSearch builds Google and Google News searches around software version lifespans.

VerSearch uses the endoflife.date API to get the end-of-life data for over 200 software products (operating systems, browsers, etc) and uses that information to create Google and Google News searches across the length of that version's lifespan. Here's how it works:

1. Use the first menu to choose a software product for search.

You can get a full list at endoflife.date, or just use the autocomplete function of the drop-down menu.

2. You'll get a dropdown menu of versions for that software product along with the release date.

Choose a version and click the "Fetch Cycle Details" button.

3. You'll get information about that version's launch date and end-of-life date

(if no EOL date is available it defaults to today's date.) You also get links to time-bounded Google News and Google searches for the software product, with the time boundaries being the lifespan of that product's version. The search query is just the name of the product; expand it as you explore the search results.

If you want an example, use ios or mediawiki (the autocomplete is lowercase which is why I'm making the examples lowercase.)

Start Here






No Shop Sherlock Tries to Eliminate Cruft From Your Search Results.

No Shop Sherlock is a Google search filter that offers a few different ways to eliminate clutter in your search results. Use the dropdown menu to choose one of four different filters:

GENERAL CRUFT - Removes eBay, Amazon, Twitter, Facebook, Walmart, Instagram, etc. Also attempts to use url patterns to remove some references to ecommerce and shopping.

ONLINE BOOKSTORES - Removes Amazon, Barnes And Noble, Abebooks, Thriftbooks, etc. Also removes a few ecommerce sites that have a lot of books.

SOCIAL MEDIA - Removes Facebook, Twitter, Pinterest, Reddit, LinkedIn, TikTok, Quora, etc.

VIDEO SITES - Removes YouTube, Vimeo, Twitch, TikTok, Instagram, and Dailymotion.

Just enter your query, choose a filter, and hit the Go button. Your completed search URL will open in a new tab. Note that the eliminated keywords and sites use up much of Google's query limit, so please keep your query to ten words or less.

Did you know that Google's search results look different depending on how you order your query words?

It's true! Do a Google search for

hairstyles 1970s "hot rollers"

Now do a search for

"hot rollers" hairstyles 1970s

Do you see the difference?

Shuffle Search takes a 2-, 3-, or 4-word query and creates a list of all possible orders for those query words, and then generates clickable Google searches for all of them.

Two words: two possible searches

Three words: six possible searches

Four words: 24 possible searches

(and now you know why the limit is four words: there are 120 possible combinations for a five-word query.)











Did you know that you can weight Google searches by repeating a search term?

Try searching for farm animals cows chickens pigs and then search for farm animals cows chickens pigs pigs pigs pigs pigs pigs pigs pigs pigs pigs pigs. You'll see the second search is much more pig-oriented, though its results more generally ascertain to farming and farm animals.

Google's query limit is 32 words. Sinker Search takes a query and a word to be emphasized in the search - the "sinker" that weights your results a certain way- and builds a Google search containing the query once and the sinker term as many times it will fit inside Google's query limit. You also have the option to do a "half-sink," where the sinker term is used half as often. This is useful when your sinker term is completely overwhelming your search results.

When you click the "Sink that search" button, your query will open in a new tab.


. This is the part of the query you want the most emphasized. It can be a multi-word single term (like "modern art") but it must be a single term.





Carl is a longtime Patreon, and I really appreciate him!
He gave me an idea for a name search tool.

Carl's Name Net takes a name and optional keywords, generates a set of name variants (for "John Paul Smith" you'd get John Smith, Smith John Paul, JP Smith, etc) and builds search URLs for Google, Google Books, Google Scholar, and Internet Archive. For the Google searches, it creates two sets of searches: one for common name variants, and one for uncommon. (If you don't specify a middle name, you'll only get one set of name searches for each resource.) For the Internet Archive, every name gets its own search URL.

The keyword option is for filtering out irrelevant results. If you find you're getting too much junk in your results, add general keywords here. Keep them general: if the person you're searching lived in Paris, consider France as a potential keyword. If they taught at Notre Dame, consider adding professor before adding Notre Dame.








Anything you add here will be added to the end of the name query, hopefully to help filter out useless results.
Don't get too specific; if you're looking for physicists, add physics. If you're looking for someone who lived in Berlin their whole life, add Germany.



Find Uncommon Name Orders While Excluding Popular ones

Most names in English are expressed in news articles and other places like this: Firstname Lastname, or possibly Firstname Middlename Lastname. However, it isn't the only pattern used when writing a name.

TABNS takes a name and generates a Google search that searches for the name in reverse order (Lastname Firstname) and specifically excludes the most common expression of firstname lastname.

It changes the tenor of the search results completely, surfacing many more legal- and data- based results. Try it yourself.

If you find yourself getting too many irrelevant results, use the keywords field to add a few single-word modifiers that describe the person broadly. For example, for Jimmy Carter you might enter President and Georgia. For Annie Lennox you might enter music and Scotland.

In addition to flipping the name search, the Anti-Bullseye also eliminates a lot of big sites that tend to clutter up search results, including Pinterest, eBay, Amazon, LinkedIn, Facebook, and AbeBooks.









Google/Bing News Query Builder

Recently Google's been experimenting with not showing a News filter in its search results. I use Google News a lot and I didn't want to be subject to the whims of more experiments, so I made this query builder.

Enter your search query. You can use special syntax, but since this tool generates search URLs for both Bing News and Google News, you'll break one of the URL sets unless you're using a syntax both search engines have in common. Click the button and a set of search URLs for Google News and Bing News appear. Each URL opens in a new tab, so you can use this list as a jumping-off point for exploring



Super Edu Search takes higher education institution information from the Department of Education and applies it to a Google search. Did you ever want to search the Web space of all the public universities in Indiana? Or all the HBCUs in the country? Or maybe all the Baptist institutions in Texas? Now you can.

Google has a query limit of 32 words, so queries that return more than 15 results will be divided into multiple search URLs. Feel free to use any Google special syntax except site:.

Want to learn more about how Super Edu Search works? Check out this ResearchBuzz article!





HBCU and "Predominantly-Black Institutions" are NOT the same thing and in my testing showed two entirely-different sets of results.