Consulting engagements that work: a case study with Rhodes College

Welcome to a new episode of On the Air with Palantir, a long-form podcast by palantir.net where we go in-depth on topics related to the business of web design and development. It’s June 2016 and this is episode #5. In this episode Account Manager Allison Manley is joined by our client Justin McGregor from Rhodes College. Allison caught up with Justin at DrupalCon in New Orleans last month, and spoke with him about how his school has implemented Drupal, how we worked together, and how it’s been going since.

We'll be back next Tuesday with another episode of the Secret Sauce and a new installment of our long-form interview podcast On the Air With Palantir next month, but for now subscribe to all of our episodes over on iTunes.

TRANSCRIPT

Allison Manley [AM]: Welcome to On the Air with Palantir, a podcast by palantir.net where we go in-depth on topics related to the business of web design and development. It’s June 2016 and this is episode #5.

I’m Allison Manley, an Account Manager, and today my guest is a client of ours named Justin McGregor from Rhodes College. I caught Justin at DrupalCon in New Orleans last month, and spoke with him about how his school has implemented Drupal and how it’s been going.

I am here with Justin McGregor from Rhodes College. How are you?

Justin McGregor [JM]: Doing well, doing well. 

AM: We’re in the last legs of DrupalCon 2016, and we’re in the busy lobby of the conference hall in New Orleans. So there’s a little bit of noise behind us, but we will plow through. So tell me a bit about Rhodes College.

JM: Who we are and what we do? We are a small liberal arts college nestled right in the middle of Memphis, Tennessee. If you look at a map of Memphis from above, there’s a big ring road that runs at the side, and if you throw a pin in the middle, that’s us [laughs]. We are 2000 students, roughly, 300-odd employees, traditional liberal arts curriculum covering everything from pre-law, pre-med, that sort of thing, all the way down to the study of the classics. We have a Greek and Roman studies department, and right across the quad we have people that are doing pediatric oncology at St. Jude Children’s Research Hospital. So it’s literally all over the map. And I get to support them all [laughs].

AM: Wow. So your college is on Drupal. How did you come to choose Drupal in the beginning? 

JM: I’ve personally been an open-source advocate for a very long time. I’ve been in higher ed web work for going on 16 years now, and I’ve worked with a lot of different CMSs, none of which I could really ever evangelize for. They were really good for what they were, but, you know. At my last school I had evaluated Drupal 6 early on, but our CTO was very anti-open source. But I kind of fell in love with it then, and I’m like, one day I’m going to come back, I’m going to be in the right position at the right school to do this, Years elapsed and D7 had come to be a really mature solution, and I got the job at Rhodes. We were coming off of an aged open-text solution and also Sharepoint 2010 pointing internally. Both solutions were long in the tooth and needed to be replaced, and when I came on board, I’m like, yeah, cool, let’s do this thing, but one of my caveats was, we have to give serious consideration to open source and to Drupal specifically. And so during our CMS roadshow we looked at the two leading proprietary higher ed solutions and also to Drupal vendors for hosting and DevOps. And it became clear really early on that for our use case, Drupal was the only solution.

AM: Well, great. Thank you for choosing Drupal!

JM: I’m glad to be here, believe me.

AM: So what does your internal team look like? What’s the composition?

JM: It’s largely me. I am the only developer on staff. I work in the communications department, which actually reports to the dean of admission. So from a business perspective I work for our sales team. That being said, I do have an interactive technology manager, who is my liaison to our external services alumni and development departments. And while he’s not a Drupalist, he’s in the guts of the thing every day, doing work on the site in one capacity or another. I’m also lucky enough to have a handful of student workers, including one third-year computer science student, and while they’re contractually limited to only ten hours a week they are a massive help. And also I found out just before I left to come here that one of our vacant positions may be reclassified as a developer. If anyone’s looking for a Drupal development position in Memphis, Tennessee – just saying it’s a possibility [laughs].

AM: So you did hire Palantir, just to be transparent about things. You hired us for consulting a few times a week to support your team. What was it that you needed from us? What was it you needed to complete the project?

JM: Okay, there were two projects. Let’s do them chronologically. The first one was, we’re relaunching our flagship .edu, and while I’ve been in web work for a long time, I was new to the practice of building a Drupal site. And so I knew what I wanted to accomplish, but I would go to Contrib and, here’s all the modules that are available, I could go do this by modifying a template file, or any number of things. But I needed best practices, I needed best solutions. And you can go watch training videos all day long, but they’re around the piece of technology or the specific use case that may or may not actually be the use case you’re dealing with at the time. So having somebody – the structure that I loved was, we would start the week with ‘here’s the problem of the week’. Here’s the piece of functionality that I’m going to be building. Let’s talk through all the possibilities for how this problem could be solved, and arrive at what the best practice is for this use case. And then I’d take a couple of days, get in there, work on it, build it out, and at the end of the week – nine times out of ten, it was done, but if it wasn’t, we could come back and say, here’s the specific problems I ran into, how do we work through that? The metaphor I kept using was, we ate the elephant one bite at a time [laughs]. I had six content types, a whole ton of media assets, and, ooh, I think we ended up moving about 7000 pages, a piece at a time. We took this fairly massive implementation over the course of a couple of months, and built out the framework to handle it all, and just started shovelling in the content.

AM: And what was the second project?

JM: So I said a minute ago that we had the open-text that was the CMS for the public-facing part of our site at the time, and also an internal SharePoint 2010 set of publishing sites. So support for that is going away, and we needed a solution, and I’m like, you know what, I’m not standing up another CMS for this, let’s just go multi-site and do it all in Drupal. But then we have the challenge of standing up branded sites in a hurry for every little department, grant, professor, whatever, that had had a SharePoint 2010 publishing site before this. And so what we worked through was, first of all, building a road-specific installation profile of Drupal, so that out of the box, all of my content types were there, all of the branding was there – there’s the branding they can change and the branding they can’t change, as a site owner. And also the mechanism for site ownership and how that’s going to work. And the second part of that was to automate a good chunk of the deployment, so I can do from the handful of things that I have to do in my DevOps environment to a functioning Drupal site. The last one I set up took me about five minutes, which is not a bad way to go.

AM: I don’t know if I’ll have time to do that today, boss, it took me five whole minutes [laughs]. So what were your goals for your site?

JM: To drive recruitment, more than anything. We need students. More than that, though, higher ed is generally in the position of having – as much as we say that the mantra of the site is to drive recruitment, and we say that over and over, we have parents, we have alumni, we have the colleagues of our tenured faculty, on and on. Researchers that are coming because of the disciplines we teach and the research that we do. Researchers from around the country want to come see the research that’s being done here, to be able to collaborate, all of this stuff. So ease of discoverability of whatever piece of content that’s relevant to that audience – now that we’ve been in Drupal and been in production for a year and change, we’re really starting to look seriously at personalization. It’s sort of the standard higher ed model to have the audience navigation across the top – you’re a parent, you’re an alumni, you’re a current student, yada yadda yadda. But once you’re in the guts of the site, that just starts to bleed away, you know? And so being able to contextualize information based on what we know about the person coming in is steadily going to become more and more important to us. Part of that to be handled through Drupal and part of that through CRM integrations, with both Salesforce and an admissions-specific product called Slate. 

AM: So personalization is next down the road. Excellent. So what was your working relationship with Palantir? What did it look like on a day-to-day, week-to-week basis?

JM: So both of the consulting setups were more or less the same. We identified an hour early in the week where we could, you know, bang around what this week’s problem was, and then an hour later in the week, how did it go, staging for next week’s problems, or if something didn’t get done because I had a roadblock or whatever. Both of the guys I worked with were fantastic. Even if we weren’t on a call I could send them an email any time I wanted to, and hear back pretty darn quickly. But a lot of times I would save stuff for the call because it’s just something you need to talk through, or screen share or something. It was nice to have somebody who’s done a lot of Drupal deployments, at the front of the week, to say ‘this is what, given your current level of expertise, you can reasonably expect to get done this week without just absolutely killing yourself’. And to do that knowledge transfer early in the week that says, okay, here’s what you’re going to need to know, if you’re not hip to this, go read this, go watch these YouTube videos, whatever. Then set to the task and we’ll wrap up on the back end. Having somebody who was really expert in these sorts of Drupal deployments helping set your agenda, because I know my goals, I know my organization, but I need to know what is really realistic to do in the product in a given span of time, you know?

AM: All right, so let’s start with the bad stuff, the obstacles. What roadblocks did you run into during the course of the project, and how were we able to help you remove those roadblocks?

JM: I’m not new to development, I’ve been doing web work for a while, but there are sort of Drupal-specific things that, when you read through the documentation, or at least when I read through the documentation, like the hook system – they seem to work fine in the documentation, but then when you get into the guts of the system, you think, that doesn’t seem like what I just read, or doesn’t behave in the expected way. I had done a lot of front-end development prior to coming to Rhodes, and the Drupal templating system is something wholly different from anything I’d encountered before. I knew how I needed the site to behave, I knew what I needed it to look like, I knew how I needed it to respond, and all of that, but figuring out, do I do this from stacking a bunch of modules in order to handle fences, for example, which I ended up relying on a lot, to sort of get the markup back down to something reasonable and something I can work with and go from there. Or do I just dive off into the PHP and let’s make template files for everything, and that way I’m programmatically controlling markup. We had to come up with a strategy pretty early on and say, okay, for the sake of anybody who ever has to follow in my footsteps, let’s find a solution we like, and move forward with, that’s how we’re going to work.

AM: So what was the biggest win over the duration of the project? I don’t know if you have one win per project, or…

JM: So many. Seriously, I say that about the theming as a roadblock, but overall, both of these projects from an institutional perspective have been a resounding success. For academics not to complain about something is actually fairly rare [laughs]. That’s not to say that there weren’t people who take exception to, say, font choices, or, is that really the institutional red, it needs to be a little richer – you know, that sort of nitpicky ‘I don’t like this element of the design so I don’t like all of the design’. Overall, oh gosh, as of last week I think we’re 16 sites in, that we’ve launched so far. And yeah, all the designers have been very very happy with the end product, with the authoring experience. I’ve had some requests for new features and I like that my users are passionate enough to say, hey, this is great but here’s how we could make it even better, and to work through these things with me. I’ve got a handful of new content types that people have suggested – as I’ve been rolling out these little multi-site instances, I think that every last one of them could be used across the enterprise. And it’s great to have people that are willing to work with me on this sort of stuff, to come up with these ideas – not just for them. Probably one of the nicer things about being in a small liberal arts college is that they are mindful of the impact any change can have all the way across the organization, because they eat lunch with these people every day, you know? [laughs] So they’ve all been really successful.

AM: Fantastic. So the next step you’ve mentioned for yourself, because you don’t really have a team [laughs], moving forward is the personalization piece. Anything else?

JM: Well, part of the reason we selected Drupal to begin with was that it allowed us the flexibility to not deploy a site and then sort of be stuck with it forever. I mean, unlike some of the proprietaries, you get the tools that come out of the box and that’s kind of the end of it. You take one of their templates, you skin it your way, and you’re done. We’re looking at some design improvements. We did a complete redesign, this wasn’t just a move of an old design, we started from scratch and rebuilt. So there were design elements that have really worked well for us, and some things that, to use the industry jargon, aren’t converting the way we’d like them to. We’re not getting the traffic draw for some of the elements that the real estate they’re on deserves. So we’re going to take some time with the design team this summer and look at redesigning certain elements, and – I love that my templating architecture is flexible enough that there’s going to be no problem to just drop in there and, it’s all the same entities, it’s all the same data, we’re just presenting it differently. It looks like it’s going to be a fairly painless process.

Also, since we’re at DrupalCon, I’m going to mention this. I’ve been at several sessions about paragraphs over the course of the last couple of days, and when I was at DrupalCon LA I went to a session and I was like, hmm, neat idea, maybe when it matures a little bit more – well, apparently it has matured a lot over the course of the last year. Because some of the things I’ve seen people do with paragraphs here are really impressive. So I’m sort of starting to daydream about some of the tools that I may be able to give my content creators, to do a lot more complicated and a lot more interesting things than they are now. There are certain things, the demand of the site and the demand of the brand, that will require us to leave some things static, but I want to give them as much creativity and as much flexibility as I can, to really make their content sing. 

AM: Are you going to be able to add to your staff?

JM: We’re trying to get a position for another developer, which hopefully will allow me to pull out and do a little more high-level stuff. We’ve also been steadily training more and more of our communications staff to work directly in the CMS and not rely on me or Nick or the student workers to do the layouts and content for them. And again, so far that’s gone really well. Like I said, the reason that I liked what I was seeing of paragraphs and a few of the other sessions is that, as much as I’m worried about user experience for our audience, I’m also worried about user experience for my content creators. I want them to want to work in the CMS, you know? And anything I can do to improve that situation for them – out of the box Drupal’s a great CMS to work with, but there are always ways to make it better. I’m always on the lookout for tools to help make their work better.

AM: Isn’t that the thing about the Web, though – you can always make it better. Tomorrow’s another day [laughs].

JM: Exactly.

AM: Well, thank you, Justin. I appreciate you taking the time, I know you’re fried – we’re all fried on the last day of DrupalCon – and there’s been a lot of knowledge and alcohol shared [laughs]. So I know everyone’s ready to relax a bit.

JM: Next stop for me is the streetcar, I’m going to go to the other end of the French Market and start tchotchke shopping for the kids and hit three or four bars on the way back to the hotel, that sort of thing [laughs].

AM: Well, hopefully you’ll join us tonight at trivia night, and then at sprints tomorrow.

JM: Don’t know, I have family here in town so I have some obligations there. But I’d like to put in a hour or so at the sprints and see how it goes. 

AM: Well, thank you so much. I appreciate the time.

JM: Thank you!

AM: Thank you so much for listening. If you want to hear more episodes of On the Air with Palantir, make sure to subscribe on our website at palantir.net. There you can also read our blog and see our work! Each of these episodes is also available on iTunes. And of course you can also follow us on twitter at @palantir. Thanks for listening!