Northwestern Research Portal
Frequently Asked Questions
You're looking at a rebuild of the Research Portal. This page explains what it is, why we're doing it, what changes for you, and where your feedback would help most.
What is this?
This is a ground-up rebuild of the Research Portal you use today. The purpose is unchanged: one place to look up sponsored and non-sponsored project financials, personnel and salary data, pending proposals, and IRB / IACUC protocol information.
It is not replacing the current portal yet. Think of it as the same job, rebuilt on a foundation we can actually maintain and extend.
Why rebuild it instead of improving what we have?
The portal you use today works, but how it was built had become the bottleneck. Five things drove the decision:
The rebuild puts the portal on the same platform as the rest of our applications, which means the people who maintain it know it well, changes ship in days rather than months, and problems surface before you find them.
The part that matters for what you ask for next. A lot of what the old portal had to build by hand, this platform already does. Some of the differences you'll notice took roughly one line of configuration each:
- Turning extra columns on and off in a table
- Paginating a long list instead of scrolling it forever
- The notification telling you an export is ready
The same is true of the plumbing you never see — sign-in, permissions, and a full audit trail of who changed what are all shared, maintained components rather than code we wrote and now have to babysit. In the old stack, each of those was a custom build, which is a large part of why small requests turned into long projects.
Not everything will be quick — anything that needs a new Cognos report still depends on that report existing. But the floor is much lower than it was, so “can you add…” is now far more often a conversation about configuration than about construction. That's the main reason it's worth telling us what you're missing.
What's actually different?
The changes you're most likely to notice:
| Portal today | Rebuilt portal | |
|---|---|---|
| Signing in | Custom-built login that would sometimes drop you on an “Oops” page for no apparent reason, with nothing telling you why or what to do next. | Northwestern Online Passport and Duo through the University's standard authentication library — the same sign-in as other University systems. If sign-in does fail, you should get a real reason instead of a dead end. |
| Long tables | Infinite scroll — keep scrolling and hope the rest loads, with no sense of how much is left. | Most tables paginate, so you can see how many rows there are and jump around. The investigator list is the one that still scrolls continuously. |
| Exports | One bulk export for the whole project. If you only wanted the payments table, you still got everything. | The bulk project workbook is still there, and most individual tables now have their own export button too — payments, cost share, expense categories, transaction detail, and the financial overview. |
| Columns you see | A fixed set. Anything else the Cognos report returned was dropped before it reached you. | Same columns as today by default, plus a toggle to turn on additional columns the report actually returns. Nothing is hidden from you that Cognos gives us. |
| Caching | Only some reports were cached, so performance was inconsistent — one page felt fine and the next one didn't. | Every Cognos report goes through the same caching layer. That should make performance both better and more predictable — and it's something we'd like you to sanity-check. |
| Confidence in changes | No automated test coverage. | An automated suite runs on every change, plus error tracking that alerts us when something breaks in production. |
What's improved?
Some improvements don't fit a neat before-and-after, but you should feel them:
- A snappier interface. Screens should feel quicker and less janky to move around in — fewer full-page reloads, less waiting on a spinner before you can do anything.
- Exports that are fast and consistent. A single table's export streams straight down as a CSV with no waiting. The full project workbook is built in the background and notifies you when it's ready, so a big project never ties up your browser. Either way the output has the same predictable shape every time.
- Field definitions that are actually current. The tooltip definitions exist today but stopped updating years ago. They now sync on a schedule from Northwestern's official business glossary.
- You can tell how fresh the data is. Each project page shows when the data warehouse behind it last refreshed, and whether that refresh succeeded.
What stays exactly the same?
The data. Every financial, personnel, and protocol figure still comes from the same Cognos reports the current portal uses. We are not recalculating, re-deriving, or reinterpreting anything. If a number looks wrong here and right in Cognos, that's a bug on our side and we want to hear about it.
Who can see what. Researchers see their own projects, salary data, and protocols. Staff see the full investigator list and all project data. Your access is determined at sign-in from the same source as today.
What do we need feedback on?
Rebuilding from the outside in only works if the people who use the portal every day tell us where it falls short. Five things, roughly in order of how much they'd help:
Do the numbers match?
This is the single most valuable thing you can do. Pull up a project you know cold and compare it against the current portal or Cognos directly. Report any difference, including ones that look like rounding.
Is anything you rely on missing?
A column, a filter, a report, a total, a way of sorting. If a step in your routine has no home here yet, that's the gap we most need named.
Does it hold up for real work?
Try to complete an actual task end to end rather than clicking around. Where did you get stuck, backtrack, or have to guess what a label meant?
What's slow?
Tell us which page and roughly which project. Some screens pull a lot of Cognos data and we'd rather tune the ones you actually wait on.
Did a design change help or hurt?
Plenty of small layout and presentation choices differ from the current portal. Some were deliberate; some are just how the rebuild landed. If something reads worse than it used to, say so — we won't know which is which unless you tell us.
Access & timing
Do I need to request access?
Your access is granted automatically at sign-in based on the same authorization data the current portal uses. If you can use the portal today, you should be able to sign in here. If you sign in and are told you don't have access, that's worth reporting — it may be a genuine authorization issue rather than a decision.
Can I break anything by clicking around?
No. Everything financial and research-related is read-only — the portal reads Cognos and cannot write back to it. The only things you can change are your own preferences, like favorites. Explore freely.
When does this replace the current portal?
No cutover date is set. Nothing about the current portal changes until we've heard from enough of you that we're confident this one does the job better. You'll get notice well before anything switches over.
Is this as fast and as reliable as it gets?
No — and it's worth knowing why. Today the portal asks Cognos for a report the moment you open a page, then reuses that answer for a few hours. Two consequences you may run into: whoever opens a given page first does the waiting, and when Cognos itself is slow or down you get a Report Unavailable message instead of numbers.
An idea we've had since the start of this project — and one this platform makes realistic — is to turn that around: sync data out of Cognos on a schedule into the portal's own storage, and serve pages from there. Pages would be consistently quick instead of quick-once-warmed, and a Cognos outage would mean slightly older numbers rather than no numbers at all.
To be clear, this is not planned or committed. It's a substantial piece of rework and no decision has been made — we may never do it. We're mentioning it because it's the kind of change this platform allows and the old one really didn't, and because telling us which pages you actually wait on is what would make the case for it one way or the other.
Who do I contact?
The portal is built and maintained by the AppDev team in ADO (Application Development and Operations), part of Northwestern IT.
This rebuild was designed and written by Tori Duguid, who is the best person to talk to while the portal is in preview — especially for anything on this page, or a question about why something works the way it does: tori.duguid@northwestern.edu.
Email or a Teams message is the way to reach us.