MMP10: Everything But the Kitchen Sink – A campus wide web redesign perspective

Kamalika Sandell 
Associate CIO, Office of Information Technology, American University


The audio for this podcast can be downloaded at http://2011.highedweb.org/presentations/MMP10.mp3


Announcer: This is one in a series of podcasts from  the HighEdWeb Conference in Austin 2011.

Kamalika Sandell: Hi, everyone. What we're going to talk about here is a case study of what we went through, American University. And hopefully I'll be able to interact a little bit through this session, get some questions, perspective from you as I uncover some of what we went through in 45 minutes right?

I am in IT, I'm American University's Associate CIO. I did not have a higher education background running higher ed technology shops before coming to American University. I come from banking and financial and I attended this session this morning and the speaker highlighted some of the difference between coming from a non higher ed., to a higher ed., and what it feels like.

And I can tell you I fully resonated with that. It is the toughest gig of all. So, this was my -I came in 2007 to American University and things were just kind of bubbling up discussions about the web redesign.

01:08

We didn't quite launch into it. And so, here is my perspective. So, for those of you who don't know where American University is we're Northwest Washington DC. It's a private Liberal Arts University about 13,000 students, very politically and socially active student body.

The campus is buzzing always which is great. But what it also means is when we're trying to do something like this, there is a lot of opinions, right? That's probably true for any campus out there, but again this is the first campus that I got used to my introduction to higher ed., IT, technology, projects, programs if you will through Americans.

So, about our redesign program, I'm going to take you through the program a little bit. Talk about some of our challenges and then hopefully be able to again take some questions as we near the end of the session.

02:03

How many of you have recently completed a web redesign? Just show off hands. OK. How many of you are going through one as we speak? OK. OK. You're having so much fun.

[Laughter]

And how of you are thinking about embarking on one? OK. So, kind of a nice mix. Our web redesign program, we did about two whole months of planning. And the actual implementation if you will lasted for little over 18 months throughout this we worked with four different consulting companies.

The first one was company that specialized in design, a marketing design company based of New York Huge Incorporated. Then we worked with a second company which was the Content Management Vender CMS Vender that we chosen. We selected common spot which is from paper thin.

03:06

Then the third was a company that -we'll talk about this a little bit. When we went into the see of content and we figure out what to do with that we have to get just one consultant from a company who helped us develop if you will our content strategy. And then the fourth is Real UTV for our virtual tour.

So those were the four companies we worked with. And the legacy that have American University had was built in probably 2000 or 2001. It was all custom coded it was written in ColdFusion and not only did we need a redesign because the design was carried through from 2000.

We also need a rearchitecture because it was custom code. And again this is not unlike many universities when they're going through this it's not just redesign, it's also re-architecturing and thinking through things.

04:01

And things have changed so much then that year 2000 when these things happened. We lost count, but we have well over a 100,000 plus old pages. But that's just two servers and we probably had eight, so we couldn't count.

We try to get a roster of what count and inventory was looking like and quickly realized that if our content strategy was sort through all that inventory and figure out what to do that was a no win situation.

So, we'll talk about that, total content publisher these were individuals who was writing web content was about 140 as we went through this phase, And they were all over campus in different units, they created 30,000 new pages when we first launch which is again maybe it was May of last year. So, in all we did 21 focus group.

05:00

Most of these were with students with alumni. There are a couple of staff, focus groups, few faculty focus groups, some respected students we got feedback from them on design. And we were trying to track how many feedback providers, you know people who are very vocal about what they like about the wireframes and dislike about the wireframe and why something just wouldn't work.

Because you've got to be kidding, this is a university, we got a lot of feedback by the way. So, that was about 250 again plus. So, that just is a flavor why were we redesigning for all good reason that non of these are new to you. We have a school of communication, they're excellent, they said, "This is ridiculous. It's so boring, it's so clumsy."

So their focus was more on the lack of modernization of our website design. But really what we had was again outdated content .

06:02

Content was not retired so you had content that conflicted with each other. We have universities in general have some slim IT staffing, we have a slim -couple of people IT shop that dealt with the web and if you needed a new site they made all the update.

So it was a very limiting way of doing work. And by enlarge the way it could really find anything if they found anything because it was conflicting sets of information and yes we had different dates for different tuition charges and different dates for payment of tuition. All kinds of things that you wouldn't want to have. And it was just difficult to find. I remember in my first year I went to a business school and I was talking to them about my financial industry background.

And at the end of the session, one student said, "Why is our website so bad?" And that was her question. And she said," You know I try to look for something about tuition payment dates and I got a date for an ice cream shop that doesn't exist on campus."

07:11

I mean you know, ridiculous examples, but again those were real examples that people, students would give us all the time. It was clumsy, you know you just didn't have a design. It was a nice blue bar with some text and lots and lots of text out there.

Media was incorporated because we had a rotating frame that had kept your images fresh, it was in a couple of places in our old website, but that was about it. So,everyone was talking about this, and this is the time I came to the university. Finally, we had an interim president, we had an interim provost. It was a lot of interim.

A lot of our senior leadership at that time that the shatter was going on was intern and the challenge with that is always, do you embark on something or do you wait for the leadership to go beyond interim and solidify before you do up redesign as big as this.

08:12

Well, finally our admissions unit bread and butter they said we're 95% American University's 95% tuition driven. So, the fear was we're actually going to lose students and there was some trends, some data that they could show that solidified that fear.

So, things started moving and we looked at this so few units that you would imagine that the way the web staff is organized in the university. Again, there is marketing3 university communications marketing, they have -they do brand, design. We have IT folks who understand deep technology, database, servers.

09:01

We have some web development folks who work to translate the requirements coming in from the campus and then work with the deep technical folks to implement them. Each school and most programs have what you would call their titles were all over, but somebody who deals with web either in the school or in the units, you would be able to find them if you look for them enough.

Now, there was no commonality between their titles or even their job description, but if you went into a unit and you said, "OK. Who does this, this in the other for your web." I go, "Oh, we go to that person." So, we were able to with some creative questions track who these web people were across campus. And they became our first group of content publisher.

So, there was tons of stuff to read up on, I think we did a decent job on reading up what makes a web redesign program successful? What are the other things you look for? Now my experience was more from of the financial sector, the for profit corporations.

10:05

So, yes I had to read up a lot on why can't you just say we're going to do this and everybody rallies. Well, that's not how higher ed., works.

[Laughter]

It was quite a learning experience. So, one of the things we knew. So, this stuff we knew going into it that you need to have very clearly state of goals what are your redesign objectives? What are you really trying to do? For something as big as this staff availability, so you got the core units, but what about all the folks around campus who need to develop content or redesign, or be able to spend time so and so.

You know, who are the people on the ground who need to participate, who are their managers who need to be OK with the time commitment that this might take from them and how do you do all those negotiations ahead of time. So, we learned that that was one of the key things we had to do the competency of the staff involved.

11:04

So, even looking at my own shop we have folks there who knew very good ColdFusion but they may not be as savvy in all the trendy tools out there and back then we do know that we were going to do common spot. Could be any tool, could be an open source which we looked at.

We needed to buy in from the staff in academic units so you know it's great and this the other thing that I learned was when you go out there everyone's complaining and you know, 'this is crazy, it's crappy, it needs to change, how can we sustain this'. Then when you go "OK, here is what we need you to do as we go through the change program."

You get a different reaction, you get, "Really? How come the university can't fund? Well, university will fund but we got to do that work and it's a ton of work. We also prepared our leadership that there's going to be decision making involved. We needed budget so typically again you know, you can do a lot of things by the side of your overflowing desk,

12:09

But this one we would need to really make sure that we got everybody together at the same time working on the same schedule somewhat. So, we needed to make sure that the budget whether it's the Dollars to buy a tool or the servers or what not. the budget for some of the consulting companies.

See, the university when we looked at the core skill sets that we had, design was not one of the core competencies. So, it was a pretty unanimous decision to get some consulting to help us with the design starting out. But then along the way we did a pretty good job of prepping our leadership that this is the kind of budget, this is the kind of timeline commitment that you're looking at. And they did agree that they were going to engage.

13:02

Bottom line it was just a ton of work. it was probably the biggest project minus our ERP. So, OK. We are off to a good start believe it or not we really had a good plan, good engagement, we stated our goals where we wanted to be, what are some of the weak links that needed to improve.

I think the engaging of the consultant with the design experience. Now if your campus has that skills set that you feel comfortable with then you've OK. But we didn't then we were talking about intern leadership.

Some units were not getting enough exposure let's say at the time we started the university communication marketing was also had a interim leader. We knew that we had an open search going on but we didn't exactly know how that would pan out.

14:01

So, we took two whole months to figure out what the vision was and what the implementation plan was going to be. Did a lot of user research for example when we have campus days when the prospective students are coming to campus, a bunch of people, bunch of staffs to go out there and got feedback.

What are the missing, what did they go through to get the information they got or in some cases they did not get. Our president did a great job of engaging and you know that made a big difference all though later on we found out engaging the cabinet is one thing, it's not everything that's needed.

We needed much more engagement from the deans, the schools, associate deans, so on and so forth. And there were a few decisions that were made up front. So, it was decided that it was going to be a big bang approach, universitywide redesign.

15:00

We will not going to do a pilot, we were not just going to state one area and move it over so it's going to be a broad overall redesign. And yes there will be an enterprise content management tool we have to buy -select buy figure out what that would mean for us.

We would have three to five special or back in the old days we used to call them killer apps, that would be the big draw for our folks to either want to engage in the redesign or just special feature is that would make it kind of cool but no more than three to five. And yes it will be fully distributed publishing so the more people we can engage to create the content, the better off we are.

So, these were early decisions. And the university did realized that we need to engage at all levels, it's a massive effort, and you know, this is the third month, everybody said, "Oh yeah we still really need it." This is the thing.

16:03

So, we went full steam ahead. And I was beginning to feel OK, well this is pretty good. I mean, who said things move slowly here, this is pretty fast. My last job before coming here was Capitol One and I was comparing -I was in this phase where I was still comparing very much oranges to pineapples, or whatever you want to call it. So, I thought wow this is very fast, this is great.

But see, then what happened was we produced the first set of deliverables, the team did. And everybody wanted to win. And that just was -we started feeling good because people are paying attention, we raised the awareness and it's all good thing.

But everybody wanted to weigh in, and not just weigh in, they held us accountable to incorporating al those great comments design choices where very very widely. And then everybody started telling us what we were not doing right.

17:02

You need more focus groups. What are you doing? You only have six focus groups, well you can't be in a focus group mode forever. And then we heard well, we're giving you feedback you need to circle back with us again. So, this infinite loop of get feedback, go do your thing, come back again, we'll give you more feedback, we were in the middle of that.

That turn was happening. And see this all hit us right about the fourth or fifth month all at the same time. Our cabinet members became very picky on real estate. We would go to cabinet meetings to present a concept or a wire frame and all you would hear is,"Where is my button." That's what they would say, "Where is my button." You don't have finance and treasury which we report to by the way. Why not?

Well, we're thinking you know, primary audience maybe they don't want to know about finance and treasury when they come to the homepage.

18:01

And then they would pick a spot on the page like, "We need to be in that part of the page. It's not good enough that you've stock us in the bottom right hand corner." I mean this is regular right? Am I saying something that just is just very unique to American?

OK. Faculty said their needs were not just represented if the primary audience for a website is student, you are disregarding the intellectual capital of the enterprise called the institutions. So, they were almost offended and our senior leadership felt that web tool was chaos.

What if you had that the use case that they true back at us over and over again was you have a situation where a faculty's tenure application has been denied. And students are lobbying for it? What if this happened? What are you going to do?

19:05

So, to take our university to accept and embrace web tool we have a feature it's called aupedia, it's very much like Wikipedia and it's original intent was very much suppose to be like Wikipedia where think about insider guides being written up by anybody in the American University community and then you can go and update it and keep it fresh.

To put this out there was a big big challenge for us because on the one hand we had a lot of excitement from folks -the students in particular or end staff. But on the other hand we were just being bombarded by our university what kind of liability are we exposing the institution to by actually having this on the university website could be anywhere didn't matter while on the university website. We did a decent job managing that debate.

20:02

And then the technical folks had a lot of philosophical debates on whether or not we should do open source or just a regular off the shelf product. And that's a topic in itself because I am in IT but it was a very interesting debate. We ended up not going open source, we chose a common spot.

Part of the reason was to leverage the skill sets that we already have common spot is somewhat ColdFusion based so that really helped us. But you know the team that chose this was about 40 or 50 people across campus. So, they were the recommended ones fair and square even from the user community.

So, that helped us through the debate. Remember how everybody said they were in and they loved it, and they needed, the moment we set each wireframes and somebody didn't like any. They said, "Well can we opt out?"

21:00

Well, not actually I take that back, they didn't say can we opt out, they said, "We are opting out." So, actually before I posed it I'm going to rephrase that. And then they said, "We have to generate content." We -you're looking at us to do this?

So, think about it. it's like sinking in the messages that were given right at the get go, right? We kept on talking about this, but every time a new message actually sinked in, it sparked the biggest debate in that week or month. And then we were on the next debate before the first debate was even offered.

So, it felt very chaotic. We had between the fourth, and the sixth, and seventh month we were stuck, it was going slow. How do you deal with all of this?

22:00

So, we did this -I don't think I have it here, but we basically further organize the program, but we did this think about open sessions in campus where, 'come and talk to us about the web'. And the president engaged his chief of staff on the project. So, the thing I go back to is the reason I think we could do it at that scale is because of leadership engagement.

Because had that not been the case, had the chief of staff at that point not been engaged, I don't think IT and marketing together alone would have been able to navigate all these debates. So, we did a number of open houses, we said, "You know you're going to have to do the work because you know what you're talking about. We can get a consultant, but they won't know what you're talking about.

They won't give your messages the way you want." However there is going to be a lot of help. This is the point when we realized that we need some consulting to draft tonight's content strategy because it's too massive and we can't go through the pages there was no earthly way.

23:09

We shouldn't even have consider that was an option. So, we really really got more organized at that point. We had four different committees, but all the times we bash committees and say that it's a waste of time and it just you know, you turn and you go to a meeting to talk about a meeting and come out, go to the next meeting to reconfirm that you needed the meeting.

This were working committees because we still were holding to our 18 month delivery. There was a committee that started looking at content and come up with how do we get the campus to write content. There was a committee that started. We did not have a web policy per se for our legacy web, we had a standard of use kind of thing.

24:00

So, started looking into the policy aspect particularly if we had a social media, the feature we call aupedia then what would that mean from a policy standpoint and engaging general council and that there was a pure technical team that was looking -working still with the campus but really trying to get through the choice of the content management system.

And we had this fourth group -remember we're going to three to five special features or killer applications. So, that was the killer apps group. They try to figure out what those three to five would be. So, they looked at a whole bunch of stuff, settled in on this aupedia.

They also settled in on profiles for staff, and student, and faculty. We did not have profiles in our old legacy web. The other was a event system, university calendar where you could have master events and special events, and go by departments and so.

25:00

Depending on let's say you're on a department page or you navigate from the department or a certain program the events page gets refreshed. So, it's nice and nifty. And it's one of the things that's most widely looked at and used. We structured the feedback process because that's the biggest turn for us.

And essentially it was going to be three concepts, two rounds and then we work with a subset of the cabinet if there were debates. But three concepts, two rounds was what we said we would follow if we needed to hit schedule. If that schedule concern was anything you wanted to go by, identify where the groups would be needed so, we shared the plan, talked about where we would bring focus groups, what the plan was because we were doing many focus groups, but the feedback coming from the campus is, so keep doing more, you haven't engaged faculty from that department or forget faculty you know, your focus groups have more undergrads not enough grads. You don't have specialty disciplines built into it.

26:04

So, you could do focus groups for a very long time, it's just how much do you need and how much can you do later? And we set a lot of expectations, so we said not all feedback will be in because think about it, you're sitting in a room, doesn't matter how many people you have. You have conflicting feedback, you can't put all the feedback in, there's no way.

You can't even sometimes wait through the sea of feedback too. It's what you're doing to get a feel for how you would start. So, what we start with is not going to be the web redesign for the next five years to come or even a year. So, this whole concept that design evolves with users.

We got to put something out there that's going to inform us how to make it better to do. In number of focus groups and keep going at it, it's you know you want enough of it so you're not just completely missing the mark, but you can do it forever because users will tell us analytics will tell us what's being used, what's not being used.

27:10

So, it's not a do it and forget about it at all. So, that was the -the other thing web tool will not be wild west. So, user generated feedback and getting users to put their thoughts in doesn't mean we're back to the wild west

. And tool choice is important, we all keep hearing about this, but it's not the end game, it's what you do with the tool that really matters. Opt out is an exception and again, yes there is no way you can say to anybody that this is not a lot of work. But that's just what it is, that's what you've got to do to get it over to the other side.

And new realizations, we realized that -remember when we started we knew that design was not necessarily American University score competency, not across the board at least.

28:00

We also realized that content is king and content is queen, that's being overused, but it's very true. One of our faculty in the school of communications uses it all the time. And she's absolutely right and she also said, "Who's going to write for the web?

Do the people know how to write for the web? What kind of training do you have, Annie Eisman, she's a very esteemed faculty. And we realized when outdoor our content publishers and we knew who they were, but they didn't necessarily have any uniform title, there was not any uniformity set in terms of what their job descriptions are, or what the expectations of their job descriptions were.

There were just the web people in the units. And he told us that, 'you've got to teach them web writing. So, she at our request did a whole set of training simply on web writing. And these trainings were run in batches, I think she did it over a period of three months.

29:01

Very structured, great training, even I took it and you know she walked through what people looked for when they go to the web and stuff that. Even my IT folks took it because if we don't know what we're building at four then you kind of miss out on that fine element of, 'oh, so what?' You tune yourself to some of the things your users are saying.

So, that was a big investment from a time standpoint and again our faculty came through to develop these training. We needed our cabinet to understand what we're talking about because think about it, you go say,"Web writing is not just -web writing is a thing." You've got to maybe focus a little bit on it.

They didn't appreciate this at all. So, Annie Eisman, again our faculty she did a one hour session with the whole cabinet to talk about why is web writing different. Why is it unique? And why our old content wasn't going to cut it.

30:01

So, she reconfirmed the fact that you better off rewriting the content because here is all the problems with your old content. Don't try to bring it along, don't try to cut and paste, just rewrite and then you design. Distributed publishing sounded great, the design to support it, the ongoing training to support it, we needed to think through that.

It was not give the tool to users and they will be able to learn and work in it, it was much more than that. Multiple venders, the coordination management was painful sometimes. And creating content was definitely in always a one time activity and neither was the training that we rolled out for three months that have current ongoing component to it. Remember when we started this we had multiple intern leaders or president approvals, the communications marketing lead, the development fund raising, they were all intern, American was going through a tough time right before that, I don't know how many of you followed American's history.

31:05

However, so every time we had a leader change they brought new opinions on what they saw worked at their campuses and they wanted to rethink all the decisions because they didn't realize what the rush was. And again there was no rush per se, but you want to be a well managed program, you could turn with this thing for a really long time.

So, just the multiple engagements of every leader who came in to have them understand what was going on. And I go back to the addition of the president's chief of staff to the steering committee was a huge deal because he made sure when we engage in these new leader bringing them up to speed sessions, he would always be there and that made a big difference. So, we lost a lot of time, and budget was running out.

32:01

So you know, when they finally chose the content management tool, designs were shaping up so then the question was just build the templates and turn it on and I didn't know how to do that. There was going to be a need to stand up the basic architecture in the first place, set up the server, is that decisions around virtual servers and architecture.

There was going to be templates that need to be developed and our team did necessarily -well, did not know the tool. They were excellent technical staff, but they couldn't just turn it on like that. So, we looked at the massive sea of what need to be turned on and developed in this special features and applications, and it was going to take months. The timeline slipped somewhat, but not hugely.

So, this was the first area where we introduced agile methodology.

33:03

If I had plan fully walked into it I would not have said, "You know this is not the project where you want to introduce your campus to how to work with agile pieces of coat and develop." But essentially what I said was if you look at everything that's to be -what's a good example? OK.

So, let's say we had new stories, and profiles and success stories, OK. These were some of the elements that were being built. So, homepage would have dynamic tagging and some of these would automatically come up. Instead of starting with homepage, we built out the elements first build out the new story component the success story component. And then we would test it and give it to the publishers. And they would populate and write.

My idea was they would write up new stories and success stories. So, we build components, we throw it out there, they use the component to write the content. And the homepage for example at either a school level or program level comes towards the end, then they kind of bring it all together.

34:09

We were very successful, but not without payback. It's difficult to envision design and content. Not to envision -it's difficult to realize design and content like that. I had them do bottom up approach, they wanted to do up down, we met them in the middle somewhere. So, that had it's own challenges, but it made the turnaround fast.

So the entire development of these was about four and a half months. And then we had a phase to pull it all together, but when the publishers looked at all the content together, they did change some. Now they were very appreciative because as things started happening, they could see what they talked about and what they were engaged in be realized.

But it was painful for them because some rework was involved. Now, if you ask them the percent of rework, the percent of what they did and the time, it was very insignificant.

35:04

So, they were still happy. Involved faculty, OK I talked about this. Lots and lots of testing, the biggest multiple rounds of performance testing. So in my past experience this is the biggest area wherein other implementations not at AU, we've had a lot of challenges as you put everything together and this thing is very slow, it doesn't render quickly. So, we had a lot of performance testing rounds.

Our director of web, he implemented what - it was his caching custom cashing on common spot caching, and then you product in the new version. A lot of that has been incorporated, we are very thrilled about it, we don't have to have his custom caching anymore, but it really saved us.

And then we had rollout plan and then two levels of contingency plan. In the meantime, seeing all of these our leadership had been solidified, we had the president, we had provost, we had vice president of marketing and communications, and they wanted to make a big splash.

36:04

Never again will the campus have me do this, but the had the roll out date, they made T-shirts in it, have the date on it. And they wanted a launch event. Now, I'm IT, I don't like launch events. Launch events is when we're all sitting in a war room trying to figure out if everything is working fine or not.

So, I had to have levels and levels of contingency plan because they had , 'oh my god', they had media like it has no tomorrow. They made a big splash, thank god they did, it worked out very well, but that particular day in the morning, I did not wake up with a very happy feeling. And as it went closer, noon was going to be the cut off and the flipping, the switch over to the new -yeah, it was quite something.

It was a major success. And I knew performance, you know if you have performance you can keep tweaking design, things that work and render decently well. However everything worked as designed, but not necessarily as needed.

37:02

We had to redesign the homepage, we had lots of feedback on some nifty things set were introduced. It was more confusing, the niftiness created confusion, we took it out. This beautiful thing that we was going to be out hallmark, aupedia it was fully redesigned in 120 days, it was just complicated for users to give feedback, yeah.

Analytics had issues, we were not getting good data at that time we rolled this out we had we have web trans and we use Google analytics right. And we had analytics in some custom elements was not happening well and then we had to do an upgrade of web-trans.

So, there were a lot of follow up project that happened after that. Performance did d great in the first four or five months and then we had to upgrade all the three servers. So, a project and many other projects.

38:01

Where are we now? This two pages and then I think that's pretty much our case study journey. So we do have a web policy, and we worked very hard. We put it -the web policy was signed in last November. I'm going to try put some resource links when I publish this.

And the web policy -there is a steering committee. We do have web governance now it's no governance out there is perfect. So again, it works some elements of it work, not everything does work. The web steering committee meets every two weeks diligently, it doesn't -yup. And it's a very invested community, it's chaired by the Vice President of Communications Marketing.

And every unit has to have an individual assigned to the steering committee that is one level below the vice president. You can delegate specific sessions, two others if you're not going to be there, but you can't delegate the whole roll over to somebody in your unit.

39:04

So, it tries to create the strategic view. There is a content publishing leads committee, so these are the individuals in the schools. Now, we did - we could not change job descriptions necessarily, but we have a roll content publishing lead and then a content publisher across campus.

These roles are defined so the content publishing lead, they meet -they used to meet every two weeks, now they're meeting every month. Then half of their session covers more analytics and web writing oriented topics, and the other half is technical topics.

There is regular training, there is three levels of just the technical training in the tools starts out with beginner, intermediate, advanced, and they're all offered online, but there's some classroom sessions too. Performance continues to be a major focus. Again we have learned that it's not something you do and forget all about it.

40:01

If you have distributed publishing it's very easy to break things. We have templates out there, but still it's easy to do a query that brings so much. We found out that the news element on a particular school page was rendering so much data, that it broke performance. It broke the website.

So, it's definitely an ongoing challenge on weighing metrics. We do annual planning one day onsite, and I have every six months with the steering committee to show them what other universities are doing and what's going on, where we are, not unlike again where other universities were taking the mobile journey, better social media integration.

We did moved to Google analytics, we have more work there to figure out how to get our VPs to tune in to the metrics. That's not the challenge, the challenge is so what? What do you do with them ? Are you really making any decisions based up of that? We've gotten people to tune in to the metrics now.

41:01

But we haven't got the making decisions of it if necessarily, so that's definitely a focus. American wants is the new branding for American University, so we're making sure the website reflects the new brand. So, it's you know the biggest fear that I have is that it's going to go back to the same elements.

Remember that little thing that had outdated content and redundant content. So, with all kinds of governance and everything, how do you make sure that someone is not duplicating content out there. I mean you can't be a cop. And that's a massive awareness raised over time things degrade.

It's still new for us right now. So, that's pretty much you know. The new stuff -this don't worry me, these are going to happen. It's doable. It's keeping the basic things in place that do continue worrying me.

42:01

I might be at time huh?OK. What questions? Go ahead.

Audience: Why you said your operation redesign within 90 days. Does that mean you did another design or another redesign on that within 90 days?

Kamalika Sandell: Yes. We brought together about three different focus groups very quickly and the feedback that we got was, we had a drawer up top which you could open and you could look at things and everybody said that was very very confusing and wasn't good at all. So, we flip that around -remember all the vice presidents and their cabinets used to say where is my button? And I want that piece of real estate so we had giving and development a really nice section and our students hated that, that was not right. They wanted to come to the homepage.

Audience: Did you use testing based on the design of your second? I can read an art design and navigation blocked into everything. And it was really great for our students, and our younger faculty, but our older faculty, alumni navigate that website and it looks terrible. So that's one thing we're trying to address is how do we appeal to both younger members and old members are going to be active.

43:27

Kamalika Sandell: We did a lot of usability testing. But if feedback conflicts, you know sometimes you just can't put all the feedback to get one massive design because. So, we reverted back and this is interesting, because again you know, if we're here next year it could be different, but at least right now, there is a whole concept of primary audience versus all audience. And we're still trying to keep to that. So, website does have a primary audience and it's a tricky concept. Go ahead.

Audience: When you launch, did you launch every single site the same day?

44:00

Kamalika Sandell: Yes. Mine has some of the research institutes and the specialized a few units like that. The only main site that we didn't launch with the big launch was library because we just needed more time to think through. It's kind of an institution by itself so the design needed more time.

Audience: You're talking about how you decided to go the bottom of it approach. Did everybody anticipated the..

Kamalika Sandell: Right. The tab down.

Audience: And then you said though that you would recommend that you do a little differently.

Kamalika Sandell: I would spend a lot of time see, I wish that right going in we have plan for agile because then I would have had more time with the campus to explain that they're not necessarily losing anything,. How this all comes together. I had not time, it was very reactionary. So, to explain to them, just focus on your news elements. Your goal by next week is to have 100 new stories written up.

45:00

Don't worry about how it's all going to dynamically linked entirely will all come together. That a big leap of faith for them to take, so it made them very uncomfortable. Now, the very vocal ones would come and sit down, and we would talk about how it would come together. So, I would have planned for it, and I would have explained to them, have more time to bring them along, it was a tough thing, but you know, it worked.

Audience: The bottom up approach.

Kamalika Sandell: It works.

Audience: But with if it's more agile and thicker as long as.

Kamalika Sandell: Much more because for us from an IT standpoint, roll out everything, would have taken a whole eight months instead, we started having the first piece in their hands in six weeks, and they started building content with it.

Moderator: Thank you very much.

[Applause]