Alabama Community College System and its Zeal to Centralize EdTech
On November 10th, the Alabama Community College System (ACCS) Chancellor and deputy CIO announced the results of three Request for Proposal (RFP) evaluations leading to the selection of Blackboard Learn Ultra as the system-wide LMS, Honorlock as the online proctoring solution, and Tutor.com as its online tutoring system. Exactly one month later on December 10th, the Chancellor announced that all three product evaluations were void and that the system would have to start over. How did this strategic EdTech initiative get so off track to cause this reversal?
Initially I covered this as an LMS story – Blackboard achieving its most significant LMS win since 2015’s University of Phoenix decision, and Instructure’s ending its remarkable streak of no Canvas departures in higher ed for a full decade. But even in the research for that post it was becoming clear that there was a bigger story than the LMS.
In order to understand this major reversal, it helps to see the bigger picture, both functionally and historically, before looking at the details.
Functional and Historical Context
Centralizing the LMS, online proctoring, and online tutoring functions was not the first part of this broader initiative. In May 2018, ACCS announced the centralization of the ERP (Student Information, Financial, HR Systems) with Ellucian Banner. The announcement captured much of the rationale for the subsequent EdTech evaluations.
Fast forward to 2020 and add in the impact of the global pandemic, which includes a heightened need for remote and online education as well as CARES Act funding, and the three EdTech product evaluations began in the late spring. But there is still more to come. ACCS is also near the selection of a system-wide CRM vendor.
Centralization in nonprofit higher ed is always a tricky move, requiring significant buy-in from the affected campus communities. It turns out that the stakes are even higher at ACCS due to its recent history. From 2002 – 2006 then Chancellor Roy Johnson pushed through several software initiatives with The Access Group for $7 million for no-bid contracts on software for adult education and GED management systems, products that didn’t exist yet. Johnson ended up pleading guilty to 15 criminal charges of conspiracy, bribery, witness tampering and money laundering, and one part of this broad guilty plea directly related to The Access Group, as described in 2008 at Alabama.com.
While this year’s situation is very different, with no allegations of impropriety that I am aware of, that history might help explain the quick move to reverse three different product evaluations. ACCS and its colleges have cultural baggage that make it even more important for colleges to have a shared buy-in to the overall centralization effort.
Three Evaluations Voided
With the current EdTech initiative, it is also important to recognize that it was not a matter of one product evaluation getting caught up in bureaucratic details. There were three RFPs involved. In the decision announced last week, Chancellor Jimmy Baker said that the new proctoring and tutoring RFPs would be issued within the next 30 days. It is unclear how ACCS will manage the LMS evaluation, but the Chancellor directed colleges to extend current contracts through December 31, 2022. This implies that the LMS decision needs much more analysis and a revised RFP process to be complete within the next year (typically there is a one-year transition period before it is safe to cancel previous contracts).
The following sections are based on interviews with anonymous sources who were involved in or familiar with the RFP process as well as online conversations triggered by my initial coverage. I attempted but was unsuccessful in getting comments or additional information from system administration. The Director of Virtual College Programs Chris Alexander, who ran the evaluation projects, declined to comment. The marketing and communications administrator stated that “Our press release will be our only statement on the Blackboard selection.” The ACCS Senior Associate Counsel David O’Brien rejected my open records request, stating “even if you were a citizen of Alabama, under Alabama law the burden of inspecting documents, identifying the documents which are the subject of the request and copying documents cannot be shifted to the public entity by the requesting party.”
A Matter of Trust
With that context in mind, let’s look at the flaws in the LMS RFP process that likely derailed the EdTech initiative. The first problem that I noticed was that there was no actual board vote or approval for these contracts. Instead, the Deputy CIO announced that the system had made three vendor selections during the November 10th board meeting.
That sounds to me like they are positioning this as one acquisition with three components tightly integrated, rather than acknowledging that there are very separate procurement processes with completely different vendors and no consistent user interface. I have trouble understanding how this process complies with the ACCS purchasing policy – a board vote in 2018 to approve Banner should not imply board approval of Blackboard or Honorlock or Tutor.com. As stated in policy 309.01:
During the November 10th board meeting, Chancellor Jimmy Baker had a word of caution based on the LMS evaluation, and this description captured much of what was wrong with the RFP process (lightly edited).
This characterization is misguided, and I believe that the Chancellor now realizes the situation, as seen in his message from last week:
I believe the reason for the significant pushback on the decision was not a simple matter of some people preferring Canvas, some preferring Learn Ultra. The matter was one of trust in the process, and this blog post comment better captures the problem at hand.
Dubious Cost Savings
Let’s look at that $2.5 million in savings number. According to my sources, what happened is that Blackboard offered a bundle of products at one price, including Learn Ultra, Collaborate, Analytics, SafeAssign, and Ally among others. The vendor proposals did not show a 5-year price differential of $2.5 million; rather, administration added in costs for Instructure Canvas, D2L Brightspace, and Learning Technologies Group (LTG)’s Open LMS (the artist formerly known as Moodlerooms) for roughly $2 million. This addition was based on the estimated five-year costs for Ally and Turnitin plagiarism detection.
The actual RFP made no mention of this functionality – there were no requirements, no product demonstrations, and no ability for Instructure, D2L, or LTG to offer a bundled offering that included accessibility or plagiarism detection software. The only requirements that touched on these areas were as follows:
Proposals should indicate their inclusion/handling of the following features: Accessibility standards (reminders for ALT tags, etc.).
Proposals must indicate their inclusion and handling of the following required features: Integration with plagiarism detection tools.
It is clear from the Chancellor’s description that a driver for the system choosing Blackboard was this bundled price savings, but roughly $2 million of those savings were based on systems not specified or competitively bid or evaluated. I have no problem if a system chooses a bundled option that saves money, especially in 2020 with its major budget challenges. But it is important that this decision is made with transparency, at least with the entire steering committee and preferably with the entire system. Especially in a system with historical baggage.
Trust and process issues, aside, I have serious doubts whether this estimate would have proven accurate. As I described in my initial post:
The cost savings assume that the full transition would have occurred before the Fall 2021 term, and that any existing contracts with Canvas and Open LMS would end by that date. This means that Instructure or LTG would have been happy to release schools from any contracts that extended beyond Summer 2021, and the cost savings assume that the migration would be successful in that very aggressive time frame.
The cost savings also assume full migration to Ally and SafeAssign. I doubt that Turnitin was willing to release all colleges under contract for its product, either.
And finally, it is important to note that the $2.5 million cost savings estimate was based on a comparison between Blackboard and Canvas, the two most-used LMS solutions in the system. It is not even clear that Blackboard had the lowest bid of the four vendors involved, with or without the $2 million adjustment.
A Matter of Risk
There is also the question of how to handle Learn Ultra. Blackboard did not bid Learn Original, and all evaluations were based on Ultra. Importantly, Learn Ultra is not the same system as Learn Original for end users (faculty and students and system administrators), and only one of the Blackboard schools in the system had already migrated to Learn Ultra for the course view. In other words, this migration would have required 96% of the system student FTE to migrate to a new system, and most schools would not have had a full academic term in the Spring to pilot the new system.
Beyond this impact on timeframe and cost savings, I was told by more than one source that one of the scoring mechanisms used to justify the decision included looking at the number of colleges (or of enrollments, I am not sure) that were already on a particular system, giving credit to Blackboard for all Learn usage, not just Learn Ultra at one campus. Assuming that the purpose of this method was to quantify the risk of changing systems, this approach is misleading if applied for all Learn campuses.
In the end, I believe the reason for this flawed and canceled set of EdTech product selection process – the “information regarding the administration of our evaluation process” – was caused by an overzealous effort at centralization without regard to transparency or fairness. It may be that Blackboard’s bundled software approach is best for ACCS, but not based on this flawed process.
What I do not know (yet) is why the online proctoring and tutoring contracts were also voided. These selections were led by the same administrative group (the office of Virtual College Programs), and perhaps the Chancellor felt that, given the ACCS historical baggage, the flawed LMS process necessitated an aggressive reaction to void all three selections. Assuming that the system fixes the flaws in the product evaluation processes, I believe that this decision to start over will pay off in the long run, despite the pain of voiding contracts and redoing RFP processes.
For now, however, there is a real lesson for other higher ed institutions. When considering centralization initiatives, even during the pandemic, it is crucial to have vendor evaluation processes that create shared buy-in and trust in the decision process.
Disclosure: Blackboard, Instructure, D2L, Moodle, and Turnitin are all current or recent subscribers to our EdTech Market Analysis service. As a consultant, I have facilitated multiple LMS evaluation processes for statewide systems of higher education.
The post Alabama Community College System and its Zeal to Centralize EdTech appeared first on Phil Hill & Associates.