I'm a Recruiter, Not a Data Scientist!

Editor's Notes

• #5: Hi everyone, thanks so much for your time today. Today, we’ll be talking about “data” in the real-world. There is a gap in higher ed data. There are those pinnacle case studies people love to talk about “Through predictive data science, we were able to identify an at-risk group, and we introduced interactive interventions that we were able to proactively reach out in a personalized way to improve outcomes by some %” - but come on. That isn’t realistic for most institutions. Today is the opposite of that, I want to talk about real-data, the real benefits, the real challenges, and ways to approach and look at your admissions data. Data is not glamourous for most in higher ed – and honestly, I don’t think it will be. Sure data can be fun and interesting, but usually it’s messy, unorganized. And we have plenty of it. We’ve all got data laying around in various states all of the place.
• #6: More granularly, I want to talk with you about the real benefits of data . Why data doesn’t have to be as complicated as we think. How you find meaning in the data. And how that all comes together…
• #7: Many of you attended our July 2017 webinar on how to improve the value of data captured by admissions. If you missed it, or if you’d like to review it again, links to the recording and slides will be available at the end of this presentation.
• #8: And the reality is… you can do things with it – even as an admissions professional. It’ll require work. But I want to show you some simple ways to approach it with some real world case studies. This webinar is going to be a little different than most, it’ll have some slides (wouldn’t be a webinar without those) but what I really want to focus on is the data, so I’ll be jumping out of the power point, and we’ll be doing some real world exploring of some higher education data.
• #9: Firstly, Higher Education is absolutely filled with data… there is NO shortage of data sysstems and sources… web analytics, clicks, links, pageviews, sessions, users, emails sent out, click rates, advertising data – impressions clicks,social media data, mentions, tweets. A lot of content for future webinars : ) 
• #10: But today, we’re going to talk primarily about the data within a CRM or SIS. Customer Relationship Manager (targetx , Ellucian Recruit, Slate, EnrollmentRX, SalesForce) Student Information Systems (Banner, Campus Management, Jenzabar)
• #11: you’re likely familiar with this type of data already – the data comes from a CRM or SIS. How many enrollments did you have last year, how many applications did you receive (whether from your own internal system or the CommonApp). And maybe you have some way to get at your prospective students – whether it’s a list of names or individual visitors.
• #12: And naturally, this type of data can be used for a variety of “analysis / reporting” which we’ll be going into details about. By looking at the data you suddenly have accountability And who has too many tasks And most importantly, data can be a form of “truth”  And these are just some of the benefits… but… the irony is… if the benefits are so clear… why aren’t colleges doing data better????? Welll……
• #13: And that’s because there are a lot of barriers… Data is hard. I hate to say it, but it is. Sure there are ways and things to make data easier… which we’ll talk about today. But here are the BIG barriers we typically see… and I’d put them in these 4 categories. Focus, accessibility, repeatability and Context
• #14: Focus: too often the focus of data is on the wrong thing – it’s one thing to focus on making your data better, but often times, the pursuit of perfection derails the project. Accept your data won’t be perfect (ever) and focus on what you CAN do now. You should be interested in PROPORTIONS, SWINGS, TRENDS NOT decimal places. All you should care about is am I 1/3 there, halfway? All the way? Data analysis is about comparison, comparing .1 to .2 isn’t going to get you far.
• #15: Along with the same lines of focus: the focus should never be on tools themselves. If your goal is to build a better house, very rarely does someone suggest buying a better hammer. The data world has a problem with shiny object syndrome, all these cool tools – we’ll definitely be talking about tools, but this should NEVER be the focus.
• #16: The focus should be around Data Accessibility and Repeatability – Accessibility: can you access the data you need? Often times, I’ve found that the problem isn’t with the data itself, it’s how you GET the data – if data is hard to get to, that’s a problem. How long does it take you to get data? And along with that, repeatability. How can you automate findings, alerts, data. Can you regularly get the answer you need? If I give you an answer once but it takes 6 months to get the answer again: that’s a problem.
• #17: And the absolute most important thing… Focus on the questions!! - What can we answer, what can we DO with that answer? If you’re answering questions with data, and you can’t do anything with the answers, I’d suggest asking better questions. Better questions always leads to better answers. Now… this is the part where I get on a soapbox… over and over again, I see colleges failing to identify the QUESTION. What is it you’re trying to answer. It turns into “Well, if we had Tableau, or this shiny object, it’d allow us to connect this. Blah blah” those things may all be true. But start with the question – what is the QUESTION that might drive change. If we had this answer, we could do X… AND THEN focus on the barriers, the accessibility, repeatability, tools, etc. No tool is going to make up the questions for you…  Questions can also provide CONTEXT – “why are you asking the question?” - If I ask you “What is our enrollment pipeline?” From what angle? Marketing? Enrollment? One exercise that I absolutely love to do with admissions teams – take out a pad of post it notes and right down all the questions you’d like to know about your students from the data… and sort them… which questions are the most important – are there ways to ask better questions?
• #18: So today, I’m going to do all the demos in a tool called the Collegis Discovery Engine – but I can show you the excel equivalents pretty easily with a couple exceptions. If you can’t tell, we’re pretty obsessed with questions here… and that’s what the CDE is built for, answering lots of questions. And the inspiration for it, came from a tool, you actually use every day.
• #19: Whether you realize it or not, you have access to a crazy amount of data from Google. Google is really really good at questions – and answering them (and also giving you easily actionable results)
• #20: I was at a conference last month… Context is in the search The answer comes fast And then I get different ways to build on the question. Think about the barriers The data isn’t perfect – it’s coming from sooo many places. The process is super repeatable and accessible. And if you aren’t get an actionable answer, ask a better question (or search). CONTEXT
• #21: So we asked ourselves… what would it look like if you could search higher ed data?
• #22: But the big question for admissions is “What is Driving Enrollment?”
• #23: So today, one of our partner – University of NORTH Alabama, has offered to let us use a small sample / subset of their data. So it’s a single program, but I think you can see how we would cut and look at that if you had a lot more data. UNA has been a collaborator with us, for the Collegis Discovery Engine – so they’re actual daily users of the tool and the platform. But again, the tool isn’t the focus…
• #24: You could do all of this in Excel if you really wanted to… The data we’re looking at is from a simple flat file of data (single table). A lot of the analysis we’re going to be doing could be done with pivot tables for instance. You may have tools available like Tableau already… But don’t worry about that today, focus on the questions at hand. And identify the important answers. but chances are, someone on your team does know Excel– or a simple google search can find you tutorials to replicate what I’m going to show you in real-time. So again… data is from a CRM or SIS, it’s a big flat file from… Customer Relationship Manager (targetx , Ellucian Recruit, Slate, EnrollmentRX, SalesForce) Student Information Systems (Banner, Campus Management, Jenzabar)
• #38: Talk about the new types of questions we’re getting from our partners… and there will ALWAYS be new questions.
• #39: Next steps…. - benefits of data Barriers We explored some data, and hopefully you’ve got some new inspiration on ways to approach your own data. Again, the focus here is questions… and when you think about Dashboards, dashboard answers questions. And good dashboards answer the most important ones. If you’re interested, we have an article the Top 10 Dashboards you need to Grow Enrollment, that outlines the important questions.
• #40: Soft CTA – if you are interested in this tool, of course, we’d be glad to talk with you about it.