Sorting out the_answers

Editor's Notes

• #2: Welcome to MLA, this is sorting out the answers session. A reminder, this is on the tech track *read blurb from Juliane*I’m Aaron Green, *intro JM & JM*, and we’re from the CMPL.
• #6: Coldstone: : Like It/Love It/Gotta Have It*mention rotator as *good*must: catalog, events*cool: social media, RSS feeds, etc…Stats from site showed *EXTREMELY* high catalog usageStats from community survey showed equation of site to PAC
• #7: Information elements: what are they? They’re whatmay going on your site, no matter how big, no matter how small.*mention juliane’s slide of the tiny-print survey results to explain why font shrinks at bottom: too many elements
• #8: Doing it right (insert joke about driving) – everyone thinks they do it correctlyCheese metaphor – You’re hosting a party for twenty or thirty people…and you ask everyone what kind of cheese they’d like. They’ll all pick a different kind (gruyere, havarti, swiss, parmesan, bleu cheese, meunster, and one weirdo who likes limburger). However, if you give them only 3 options (brie, cheddar, monterey jack), they’ll all pick one of the three.
• #9: *Talk about basic concept, but ONLY BRIEFLY. *Ask audience to look past lack of polish to underlying concept. SPACE THESE COMMENTS OUT ACROSS ALL 3 SLIDES!This is one way to mock up a website.This majestic webscape is an accurate visual representation of how I think.I thought it was amazing, and completely intuitive. I had it all thought out…every pageEveryone else…Did not.YOU are not always right. If a category or navigational structure that you are in love with does NOT test well…DITCH IT!
• #10: This is one way to mock up a website.This majestic webscape is an accurate visual representation of how I think.I thought it was amazing, and completely intuitive. Everyone else…Did not.YOU are not always right. If a category or navigational structure that you are in love with does NOT test well…DITCH IT!
• #11: This is one way to mock up a website.This majestic webscape is an accurate visual representation of how I think.I thought it was amazing, and completely intuitive. Everyone else…Did not.YOU are not always right. If a category or navigational structure that you are in love with does NOT test well…DITCH IT!
• #12: Like organizing a deck of cards. You can sort by suit, by number, by odds/evens, faces vs non, split into Euchre, etc.Now take it to the next level. This can be done with more complex information as well. If you take the time, a survey can be turned into a card sort by gathering the results together and…sorting them into categories (predefined, or otherwise).We all organize information differently. [screencap of survey]You’ve all seen websites where you can’t figure out how to get what you want, but when you show someone else, they find it in seconds. You’re looking at the information differently from the other person. But a website shouldn’t be geared for just one mode of thought. Else, you might end up with something like this: (insert Aaron’s ugly yellow/blue/green mess)(the card catalogue) – yep, the one in the wooden drawers – is an ‘old-school’
• #13: See how different the responses are? But the majority of you thought to sort XXXX into XXXX.]But it’s really hard to tally up results from a paper-based survey
• #14: Free account allows you to create 3 different card sorting projects with 30 cards per project. It will tally ten responses per project so you can see the results.
• #15: New accounts come loaded with example of open and closed sorts, which is helpful. It give you an idea of how to set up your sort properly.I’ll go into more detail on the open versus close sort shortly.
• #16: CLICK!In the free version of software you still have nice control over informationCan sort a set number of responses from the 10, example, you only want 5 responses per dayCan set a closing date for the survey
• #17: CLICK!You can customize your messages to make them more personal and professional to add a level of trust to your sortCustomize a welcome message, instructions, etc.
• #18: CLICK!Optimal Sort makes it easy. To keep your wording consistent you can just copy the text you have created to new projects.Especially helpful when going from open to closed sort.
• #19: CLICK!This is your list of elements. What you want to sort. You have 30 elements in the free version. You have unlimited elements in the subscription version.
• #20: Example: Some people may think that Book Club Information refers to starting their own book club. What we envisioned Book Club Information being was information on attending one of our book clubs. That information can be clarified when you set up the sort.
• #21: So now we’re ready to set up our exercise. Start with an open sort.Ex: In the exercise we did at the beginning, we could have given you the list of elements (all the animals), if we let you define the categories they belonged in, we might have gotten categories such as reptiles, pets, furry creatures and more.The world is your oyster when it comes to open sort.
• #22: This is an example of the open sort. The elements are the left hand side and you simply drag them to the left hand side of the page to create categories.
• #24: Example of the personalized thank you screen.
• #25: Once you’re finished, the back end of Optimal Sort give you access to all the elements and it will tell you all the categories the element was placed in.
• #26: CLICK! (6 times)129 categories were created by 26 users for 30 elements. Information overload and clearly categories aren’t truly useful.
• #27: CLICK!Coming back to cheese: As Aaron said, if you ask everyone what cheese they’d like, there will be a variety of different answers. If you come up with a small number of cheeses, everyone will find at least one they are comfortable with. This is how you go from open to closed sort. Take the best of the open sort and test effectiveness.Small number of categories that works with the scale of your website.If you use the same 30 elements, it can help to reaffirm you have picked the right categories.
• #28: This is the hard part.Here is an example of how we tried to work through the 129 categories generated by the open sort.We attempted to whittle down the categories to a manageable number of 10 and then we tried to place the elements.
• #29: Ultimately decided on 12 categories to test again with the public, knowing that we wanted to reduce some.
• #30: Example of our closed sort. Back to Optimal Sort. At this point we opted to pay for the service to get more responses.We gave users the same 25 information elements, but this time we asked them to place them in one of 12 categories. Some categories, like What’s Happening, we labels that were suggested in the Open Sort, which we liked.
• #32: This is the type of feedback you get when you use the paid version of Optimal SortFor example, 100 people decided that they would look for the Events calendar under the What’s Happening category.
• #33: You’ve got a ton of good information, what do you do with it?Your patrons are willing to help you! Initial card-sorting exercise was just posted as a link in the catalog with no advertising and people browsing just did it. Ask for help with usability testing by posting something online or in the library newsletter. Mock up your site in Photoshop or another program with basic hyperlinks to mimic the architecture of your new site
• #34: Usability testing is where your site mock up is going to come in handy.We advertised for help with usability testing in our newsletter, people e-mailed us and we made appointments with them to test the site. Had about 30 responses. People came for a 30-45 min. test (gave them CMPL notecards as a thank you for time)Have a list of tasks you’d like them to accomplish and have them click on your mock up. Tangible evidence of what works and what doesn’t.
• #35: Based on usability testing tweak your design, labels, whatever you need to changeSet up shop at reference desks and test the changes you’ve made to design with walk-ups. Also test specific demographics. We tested Parents/Children tab in story time room right after story time ended to get our specific demographic
• #36: Find It: patrons didn’t know what it meant, too many items grouped under categoryRecommendations:(Lots of variations: Top Ten, Best of…, Recommended, Recommendations, We Recommend, Suggestions)Parents: (Killed the children’s tab and though Parents would work. Card sorting told us that it might work, but usability testing told us that it was confusing)
• #37: We were trying to rename a category from our old site that was called Good Reads. It housed bibliographies, librarian recommendations and read-alikes. Since the 1999 web site, the page also expanded to include movie and music recommendations so Good Reads was no longer a good name. We had six variations on the name for one category.
• #38: Polldaddy.com allows you to post free polls and surveys on your web site. You tell them the question and choices and they generate the code for free.We were so conflicted and had differing responses from public that we decided to test the top five labels with a poll.
• #39: Children/Teens categories didn’t seem to justify the information found there.Parceled out the information to other categoriesFeedback we got from TAC was that their parents found info re: library events and the children’s page used to be designed for parents to get info so the Parents tab seemed logical.Tested it and found that most parents weren’t sure what info was supposed to be housed under this tab.
• #40: Real users: Important! Staff cannot be a patronEven if using content management system, you still need elements and categories that test well
• #41: Final site design. Seven categories. Includes everything on our Must Have list.