Categories
Technology

Silver Linings Playbook: Hard Earned Lessons from the Cloud

Session Title:  Silver Linings Playbook:  Hard Earned Lessons from the Cloud

Presenters:

  • Bob Carozzoni, Cornell University
  • Erik Lundberg, University of Washington
  • Bill Wrobleski, University of Michigan

The presenters agreed that they are all slightly insane, which is a good sign, IMO 🙂

Presentation Survey:  tinyurl.com/edu-silver

 

Session Goals:

  1. Challenge you to rethink the implications of cloud computing
  2. Provoke you to think beyond the status quo
  3. Inform you of what we’re seeing as we move forward
  4. Learn from you by asking questions

 

SLIDE:  It’s a BYOA (Bring Your Own App) kind of world:

  1. Consumers are more in contorl
  2. IT decisions are increasingly being made by non-IT pros
  3. Big, long software selction processes are a thing of the past

We generally don’t have the time to talk through the tech that touch the consumer.

 

SLIDE:  If no one follows, are we leading?

With SaaS, vendors go straight to the consumer (business).  Consumers are driving the bus.

 

SLIDE:  IT can lead in a new way, by:

  • Partnering
  • Inspiring
  • Coaching
  • Brokering
  • Enabling

 

SLIDE:  Slow Central IT means no central IT (be on the train or be under it)

Reality is that IT is never fast enough, even if you’re operating at peak efficiency and in the best possible way.

  • Concept to delivery must be faster than ever in a cloud world
  • You need to learn to move at the speed of cloud
  • A great but slow implementation is an unsuccessful implementation

 

SLIDE:  If you can’t take risks, end users will do it for you

  • Users take risks, often without even realizing it
  • No longer is risk managed through a single central decision /contract
  • We don’t have to be reckless, but we have to rethink how we look at and manage risk

Thinking about risk needs to be shifted.

 

SLIDE:  On-premise systems are too risky

  • Cloud providers survival depends on a rock solid security posture
  • They’re bigger targets, but they also make bigger security investments
  • Your IT security officer may soon be leading the charge to the cloud!  (there’s a big difference between security and compliance)

FISMA compliance may drive some ISO’s to push for the cloud

 

SLIDE:  big vendors rock (and suck)

  • Big vendors offer stability.  Also deep pockets for innovation, though not always in the direction we need
  • Small vendors are nimble, and will actually listen to higher education
  • Mastery of vendor management will become a critical IT skill

Independently we can’t influence big vendors, but they are responsive to consortia like Internet2, EDUCAUSE, etc.  While universities were cauldrons of innovation in the past, today we’re just small fish.

 

SLIDE:  Candlestick makers usually don’t invent lightbulbs

  • Transformative changes rarely come from someone immersed in operations
  • It’s hard to see over the horizon when you’re in the weeds
  • Liberate your change leaders so they can focus on change

Be careful in selection of your change agents and leaders!  People with positional authority are often NOT the people who can actually move the logjam forward.  You have to be intentional about giving people the free time and space to innovate.

 

SLIDE:  You need to be more fliexible than your cloud vendor

  • Customization creates lock-in, “version freeze,” and raises the cost of updates
  • Alter you business processes not the SaaS app
  • The days of customization are coming to an end

Using SaaS gives you economies of scale; you may need to stop thinking about app customization and start thinking about changing your business processes.  Expectations of your customers are different when using services provided by say, Google.

 

SLIDE:  Watch your Wake!

  • Cloud adoption, like any change, disrupts lives of real people
  • Find ways to support those affected by your transformation efforts
  • Help people work outside their comfort zones, but keep it outside their fear zone

The intellectual challenge is different now..instead of integration it’s a new intellection challenge.

 

SLIDE:  Experience can be a boat anchor

  • Deconstruct habitual thinking that’s based on old paradigms
  • Old  “What problem are we trying to solve?”
  • New: “What new opportunity does this present?”
  • Most of what we have learned about procurement, rsik management, project management, financial models and staffing is changing

Consumers don’t shop based on requirements, we shop based on what we want (right color, size, cost, etc.).  How we staff will change.

 

SLIDE:  Don’t Fight Redundancy

  • Cloud makes duplication more affordable
  • Feature overlap from our perspective is micro-specialization from the end-user perspective
  • Sometimes duplication is  bad, but it isn’t bad as a principle

We don’t worry about duplication in consumer products like ovens, toaster ovens, toasters, bagel toasters 🙂  Thinking about the middleware and integration points is probably a more useful exercise.

 

SLIDE:  Welcome to the Hotel California (check-in but never leave)

  • Make an exit strategy part of the selection and on-boardiing process
  • Getting your data isn’t hard, getting your meta data is
  • Architect to control key integration points, to minimize lock-in

Does your exit strategy work?  Have you ever tested it?  Automobiles are very dangerous, but have we stopped driving them?  No!  They’re far too useful.  We insure them!

Our focus is changing.

 

SLIDE:  Reviewed Summary of Survey

 

AUDIENCE QUESTIONS

Biggest risk is the network…how do you address that?  Students have multiple ways to get to a network, whether it’s 3G, wifi, etc.  Even identity is a risk if it’s housed on campus.  Even the network itself is a security consideration.

 

Have you found a good way to deal with risk incurred by click-through agreements?  We work with procurement to review P-card stuff to see what people are buying here and there.  Click-throughs have never been tested in court…yet.

 

Can we take advantage of the work done by other campuses?  YES!  Aggregation of vendor negotiations is a good thing for everyone, sort of like a “buying club.”

 

What if the negotiated contract ISN’T good enough for your counsel?  The university needs to make that decision for itself.

 

What strategies have you used to get traditional procurement folks to get over their concerns with risk?  We’ve dealt with it from a relationship perspective.  You need to become best friends with your procurement folks.  We developed a default addendum that covers every single issue that might “light up” our procurement folks.  Our addendum supersedes your contract.  This shows that we’ve heard their concerns

 

Cloud “stuff” is very much like a cafeteria where consumers pick and choose the services they want…what services do you see IT continuing to value and maintain?

 

 

 

 

Categories
Technology

How Good are your IT Services?

Title:  How Good are your IT Services?

Presenter:  Timothy M. Chester, VP for IT, University of Georgia (@accidentalcio)

Tim brings a passion around CIO and organizational performance.  “Credibility ultimately comes from how well IT services are perceived by students, faculty and staff.”  Tim will talk about his TechQual tool.  Came into position at University of Georgia as AVP, was promoted to VP.  Day-to-day relationships are what drive success, not the position.

Web site for this project:  TechQual.org

Goals and Outcomes

  • Review the context for creation of the TechQual + Survey and Tools
  • Demonstrate the linkage between customer satisfaction and IT organization credibility
  • Understand how assessment of IT outcomes can drive continuous improvement processes
  • How to design and crae a TechQual+ survey to assess IT service quality
  • …etc (slides went too fast)

TechQual grew out of LibQual (a library assessment tool).  Tim was also involved in an accreditation evaluation of a research university, where a number of focus groups and surveys were done regarding delivery of IT services.  People who responded ran the gamut:

  • Good perception
  • Meh attitude
  • Angry with services (led by one furious person)

Do you want your leadership’s perception of IT services to be driven by random individuals in each of the above groups?  No, you want to own this narrative.  Despite Tim’s engineering background (which is why he got hired into his current position), his conversations tend to be around aspirations, dreams, and gaps.

 

The Power of Analytics

Showed us radar graphs of questions to visualize IT strengths and weaknesses.  The radar graphs specifically address the following areas:

  1. Connectivity & Access
  2. Technology & Technology Services
  3. The End User Experience

A series of questions identify a) user expectations, and b) user perceptions.  Colors indicate differences between the two.  Items closer to the “hub” indicate lower priorities.  This method can be broken down by respondent constituent groups.  Each group brings their own set of priorities.  Faculty are always a standard deviation below every other respondent group.

 

Tableau Visualization

Tim gave a demonstration of the power of visualization of data using a tool named Tableau.

 

IT Services that deliver value

IT these days is often simply administering services, not actually running it.

 

The Credibility Cycle

Ellen Kizsis book – “The New CIO Leader”

Initial Credibility > Resources & Expectations > Outcomes > Results > Back to Credibility

Alternate path:  Poor results > Reduced Credibility > (Cycle of Overcommitment and Underperformance) > Diminished Authority

 

The IT Delivery Ecosystem

  • Strategic Leadership
  • assesment & Planning
  • Operational Best practices
  • Foundations

 

SWOT Analysis

Group was divided into 4 groups and went around the room identifying their own organization’s Strengths, Weaknesses, Opportunities, and Threats.  Top 5 from each was highlighted.  This was a pretty powerful exercise.

 

TechQual+ Project Outcomes

Measures that conceptualize the effective delivery and use of technology, or effective IT service outcomes, from the perspective of those outside the IT organization

A set of easy-to-use Web-based tools that allows institutions to create surveys based on the TechQual+ core instrument, communicate with respondents, and analyze survey results.

A peer database that allows institutions to make comparisons of IT service outcomes on the campus against the performance of other institutions, aggregated by Carnegie basic classification.

 

3 Core Commitments

  1. Connectivity and Access
  2. Technology and Collaboration Services
  3. Support and Training

 

Naturalistic Inquiry

 

What is a Positive Outcome?

People tend to make a positive evaluation when the IT service is:

  1. Delivered Consistently
  2. Communication is timely, relevant, and in an easy-to-understand form
  3. Increases collaboration opportunities with others

 

Navigating a TechQual+ Survey

Tim showed us what a respondent sees.

 

Higher Education TechQual+ Major Influences

SERVQUAL, an approach that conceives of service quality as a range of expectations that can be assessed by measuring three different dimensions of service

  • Minimum Expectations
  • Desired Expectations
  • Perceived Performance

 

Survey Design & Setup

Tim reviewed the layout of the TechQual+ web site and tools available to a University TechQual+ survey administrator.  This included Options, Core items, Custom Items, Other Questions, Instructions, Preview.  The system has the ability to tailor the way you communicate the message that your survey is available to respondents by a) generic link that can be used anywhere or b) a tool that lets you upload tailored respondent lists (students, faculty, staff).

 

Sampling Concepts

  • N = the entire population under study
  • n = the sample of respondents that is representative of N
  • Random Sampling = method for choosing n to ensure that n is truly representative of N

UGA always selects 25% of population each year to help avoid “survey fatigue.”  The TechQual+ web site has a downloadable tool that will do a random sampling for you based on data you provide (lname, fname, email, field1, field2, field3, etc.).

 

To Get Good Response Rates

Tim reviewed some of the e-mail communication tools built into the TechQual+

  • 4 e-mail message notifications is the sweet spot (pre-survey, survey, plus two reminders)
  • Should be personalized (salutation and signature)
  • ReplyTo e-mail should be a high-ranking person in the organization
  • Link needs to be “above the fold”

 

Peer Comparison Functions

Tim demonstrated the ability of the system to compare institutional surveys.  Need 50 valid responses in a survey in order for it to be added to the peer comparison data.

 

ETS Planning and Continuous Improvement Cycle

Data drives decision-making; this data goes into presentation “road show” that helps tell the IT story.  Monthly Status and Activity Reports are extremely important!

 

 

Categories
Technology

Change Management in Higher Education

October 15, 2013

Speaker:  Jim Russell, City University of New York

Topic:  Change Management in Higher Education

 

DISCLAIMER:  this session is not about revision control, it’s more about how we handle new products, services, hardware, etc.  Sometimes it’s called “adoption and learning.”  Assumption is that some of the attendees are not beginners…going to talk about the constructs that they found useful at City University.

 

Agenda-on-the-fly:

  • Fomalizing
  • Commitment
  • “Breaking through”
  • Change capacity (“heat map”)
  • Communication
  • Buy-in

 

Structure of Seminar:

  • Theory:  Change Models
  • Practice:  change plans, player, and potholes
  • Practice:  measuring change readiness

 

We went around the room and asked “who’s doing change management?”  People gave some “elevator speeches” about how you “sell” change management to campus constituents.  With any upgrade or change, ROI cannot be realized unless we manage the people component of the equation.  We then looked at Prosci’s 5 tenets of change management:

  1. We change for a reason
  2. Organizational change requires individual change
  3. Organizational outcomes are the collective result of individual change
  4. Change management is an enabling framework for managing the people side of change
  5. We apply change management to realize the benefits and desired outcomes of change

ADKAR

Awareness, Desire, Knowledge, Ability, Reinforcement

Based on two premises, which are sometimes overlooked:  it is people who change, not organizations.  Successful change occurs when individual change matches the stages of organizational change.

There is a problem with ADKAR though:  how do you define “Desire” in higher ed?  This is one of the most difficult to develop because change comes from the outside, or change is “not part of my job.”  In higher education, we have a problem not with leadership, but “followship.”  How do we get past this problem of followship?

Connect to your people on the cognitive and emotional levels.  Connect to the heart via the student experience.  Understand what matters to people, including their fears, hopes and anxieties.  Focus groups and meetings before and after help a lot.  Allow time for feedback and venting, acknowledge the change and difficulty.  Lead consistently toward desired objectives.  Be sure to leverage opinion leaders and use feedback loops; respond to concerns to ensure people are heard and valued.  Reach out to faculty members who complain the most (“the loyal opposition”).  When communicating, use specifics!  Also appeal to the entire brain with story-telling, imagery, personal accounts, real world analogies.  Avoid numbers, charts and graphs.  Keep in mind change saturation.

 

Kotter’s 8 Steps of Leading Change

  1. Establish a Sense of Urgency
  2. Creating the Guiding Vision
  3. Developing a Change Vision
  4. Communicating the Vision for Buy-in
  5. Empowering Broad-based Action
  6. Generating Short-term Wins
  7. Never Letting Up
  8. Incorporating Changes into the Culture

 

Basic Steps of Change Plans

  1. Identify Business Process Owner and Governance
  2. Ensure the Future State is Clearly Defined
  3. Identify Stakeholders / Identify Key Changes
  4. Assess the Requirements to Support the Changes
  5. Establish Action Plan, Communicaiton Plan and Training Plan
  6. Execute those Plans
  7. Monitor and be willing to change
  8. Shift to Continuous Process Improvement

 

Group work was a case study on a policy change related to social media.  What elements should be in a change plan for a campus social media policy?

Engagement needs to occur earlier, using shared governance groups to get the word out.  What is the objective in developing such a policy?  That’s often a missing piece in such discussions.  What about enforcement of such policies?  Need to have a better idea of how this stuff works in higher education.  What’s missing in the conversations that need to happen to make these things happen?

 

CUNY CASE STUDY

Loose federation of colleges

  • While leadership could be assessed, “followship” had a mixed history
  • Diversity of mission led to diversity of procedures
  • Some colleges believe that their processes are integral to their unique identities

Each unit needs to be guided toward “ownership” through engagement

CUNY adopted model of “change liaisons” to work with resistors:

  • Recruit influential leaders with local followings
  • Don’t be afraid of the loyal opposition- you’ll need them eventually anyway
  • Don’t be afraid to reject liasons
  • Have a dev plan for the liaisons

 

Communication Assessment in Higher Ed

Assess your history on large-scale projects and communication

  • Who is the audience?
  • how do you get to them?
  • What do you want to say?
  • When?
  • How can students help?

 

The Group Voted to Discuss this:  Assessment of Change Readiness

Categories
Student Affairs Technology

Should Co-Curricular Activities Contribute to Academic Early Warning Systems?

I was reflecting recently on some of the things that puzzle me about Student Affairs.  One of those things is related to the strategic use of web technology on my campus…I’ll get to that part in a minute.

In February 2013, Dr. Vincent Tinto visited CSUN and gave a presentation with the title “Student Success Does Not Arise By Chance.”  He had a number of very interesting things to say about student success, one of them related to academic early warning systems.  Bottom line:  the earlier you can identify a student who is having trouble, the better.  The sooner we can positively intervene, the more likely that student is to persist.  No surprise there.  He went on to talk a bit about how our friends on the academic side of the house have tools with measurable inputs to help them flag students who might fit the “at risk” category.  One measurable input includes class involvement via in-class discussion, LMS participation, or some other measurable way.  Of course, faculty are also able to tell at-risk students by changes in appearance, disruptive behavior, spotty attendance, etc.

My question to Dr. Tinto was this:  “What are the best examples you have seen of incorporating co-curricular signals into early warning systems?”  To my astonishment, his response was that he was not aware of any such systems.  This is something that I’ve been advocating for many years, but with little success.  I think that one challenge is that many of us view our own departments without respect to how they interact with all the other areas on campus (notwithstanding the high-profile collaborations that exist on every campus).  Another challenge that I think we face is that our systems tend to be built to support the needs of our own discrete processes, without considering how the information within our system may be useful to other areas on campus.  I’m sure you’re already thinking:  Student Affairs has hundreds of quantifiable indicators we could use to identify students who may be at-risk.  And of course, you’d be right.

We all know that mere participation in activities with like-minded students – whether it’s a student/academic club, a living learning community in campus housing, student government, you name it – is positively correlated with retention and student success.  Wouldn’t it be nice if, as part of a student’s co-curricular transcript, we could see what activities they’re involved with?  More importantly, could we not also use these very same indicators – or more correctly a lack of them – to proactively prevent students from falling into the at-risk category before they get anywhere near the danger zone?

One of the things I’m passionate about is using web technologies to build web applications and services that connect students to people and other services.  My department’s mission statement is simple:  build user-friendly, student-centric online services.  We’ve built plenty of stand-alone systems, but standalone systems are a bit like a computer that isn’t connected to the Internet…pretty useful for word processing, spreadsheets, and singe-player gaming, but ultimately kind of lonely and a little boring.  Start connecting to other systems, though, and things get interesting pretty fast.  For example:  if we know what a student’s major is, why not recommend a related club they can join, or expose that information to our career centers so they can automagically show them internships and job opportunities within their major?  How about tying those same internships and opportunities to our friends in Advancement and their database of successful alumni?  Could we recommend upcoming campus events that might interest students?  If we know what a student’s GPA is (and it’s high enough), why can’t we proactively recommend participation in student government when a senate seat from their college is open?  With the focus on assessment and measurement of learning outcomes, we have a rich trove of data on which to draw (most of which is sadly locked away in disparate repositories).  I’m sure you recognize the functionality I’m talking about:  it’s a simple recommendation engine.  If you use Netflix, you already “get” what this is about.  This idea gets really powerful if we can connect it to our academic warning systems…

Being around people who share the same interests and passions as we do is extremely powerful.  It fires us up, gets us excited about what we do, and helps make our lives more fulfilling.  Why wouldn’t we want to “bootstrap” these kinds of experiences for our students?  The good news is that it isn’t that hard, it just requires us to think a little beyond our immediate system and process needs, and consider how we can leverage our information in other, often unexpected areas.

What do you think?  Has somebody already done this somewhere and I’m just not aware of it?  Hit me up on “the twitter” @paulschantz or leave a comment below.  I look forward to hearing from you!

 

Categories
Student Affairs Technology Uncategorized

The Edward Snowden Affair: A Teachable Moment for Student Affairs and Higher Ed

The erosion of our collective privacy has been going on for a very long time.  Most of us are (sometimes grudgingly) comfortable with the exchange of our personal information for useful products and services.  The biggest problem most people seem to have with the revelations about the NSA’s surveillance program is that it can and does gather digital information about everyone, and can use it at any time for any reason.  The fact that a relative “schlub” in the organization can access and use that information is one of the main points Snowden’s whistleblowing meant to get across.  This got me thinking about how we use data in higher education.

In higher ed, we’ve been collecting lots of data for a long time, and we’re bound by law (FERPA, HIPAA, etc.) to protect and retain student data.  Our student information systems have detailed security policies outlining granular role-based access, aka which employees get to see which student information.  These policies are generally structured on a “need-to-know” basis.  Here’s my two-part question to the reader:  first, how many of us in higher ed have articulated a policy about the data we collect on our students and how we use it?  Second – and more importantly – how many of us have published such a policy that has been specifically drafted for our students?

Every Student Affairs professional I know wants to use data to help our students be successful.  When properly applied, it’s a boon to our profession.  It helps us determine our students’ interests so we can help them choose an appropriate degree program.  We know this reduces both time to graduation and major changes.  Data helps us identify clubs, affinity groups, and other co-curricular activities our students can participate in.  We know that co-curricular activity participation increases retention, especially among first and second year students.  In our day-to-day jobs, we regularly use student data to determine satisfactory academic progress, GPA, eligibility to vote in student elections, reporting of all kinds, and so on.  With the move toward self-service web applications, we’re increasingly presenting data to our students and shifting more decision-making responsibility onto their shoulders.  This is a great opportunity for us to educate students on how we use data to help them, while increasing transparency about their data’s use.

In my opinion, that last bit about transparency is the key element for higher education. There are no shortage of articles about “big data” tools and how organizations use them for competitive advantage (whatever that means).  However, the tools themselves don’t address the more fundamental nature of how we “connect the dots” between disparate data points.  We should inform our students how we use their information so they can make better choices.  We should teach our students how they can use this information to their advantage.  We should help students understand that they are the masters of their own data.  In the same way that Mint.com provides insight into how we spend our money, our student data should provide insight that lets students thoughtfully determine where they should spend most of their effort.

What do you think?

%d bloggers like this: