Ann Ditlow: Data Analyst at 4Front CU
Welcome to our very first edition of “A Day in the Life of a Data Analyst,” featuring the equally talented and down-to-earth Ann Ditlow, Data Analyst at 4Front CU. Ann ...
An innovative, world-class platform that includes the best elements of a data warehouse and a data lake.
It seems we can't find what you're looking for.
Hosted entirely in the cloud, the Gemineye Data Lakehouse offers credit unions and community banks the best of both worlds in a data analytics product. Combining a data warehouse and a data lake together allows for data storage in a raw format, but makes it easy to transform into something usable for analysis and research.
With business intelligence, reports, data science and analytics, and AI, the Gemineye Data Lakehouse creates an efficient and cost-effective option for community banks and credit unions.
Finally, more credit unions and community banks can leverage this critical technology to improve their customer experience.
All of the Gemineye Data Lakehouse systems run on Databricks architecture, globally recognized as a best-in-class data solution and used by Fortune-50 companies. This structure allows us to provide an unbelievably efficient, leading-edge, flexible, and low-cost single-stop-shop for credit unions and community banks.
The Gemineye Data Lakehouse operates in a single, cloud-native environment, allowing our solution to be more nimble, user-friendly, and affordable. While competitors still use on-premise or mixed-cloud solutions that are antiquated and clunky, we’ve been cloud-native from the start. Your cyber security and audit team will thank you.
Looking for more control over your data? Our innovative structure allows for precisely that. Instead of you conforming to your data analytics provider’s requirements (like member or customer definitions), the Gemineye Data Lakehouse conforms to yours.
The Gemineye Data Lakehouse runs on breakthrough, Fortune 50 architecture to create the most flexible data analytics solution available.
Welcome to our very first edition of “A Day in the Life of a Data Analyst,” featuring the equally talented and down-to-earth Ann Ditlow, Data Analyst at 4Front CU. Ann ...
The team at Gemineye is excited to announce their five-year anniversary with Quorum Federal Credit Union, a $1.1B organization headquartered in Purchase, NY. Quorum is an entirely-online credit union with ...
Gemineye (formerly The Knowlton Group) has partnered with Databricks, the world’s leading data and AI company. Gemineye’s innovative data analytics architecture, called the Gemineye Data Lakehouse, is run entirely ...
Examining the Correlation Between AI Usage and AI Sentiment in Community Financial Institutions In May and June of 2026, Gemineye conducted primary source research on AI, with the goal of examining the correlation between AI usage and AI sentiment among credit unions and community banks. 30 credit union and community bank employees from almost every business unit were surveyed, whose responses were contrasted with AI research from global leaders Gartner and Qualtrics. The results were eye-opening. A disconnect between the executive push for AI and the organizational capability to do so became clear. And your overall feelings on AI were quite surprising. What You’ll Learn from This Report How financial institutions are currently using AI The disconnect between the executive push for AI and the organizational capability to do so How community FI employees like you feel about AI in Q2 2026 High risk vs low risk AI models Why This Guide Matters for Credit Unions and Community Banks This report will help you understand where your financial institution sits within AI usage and sentiment as compared to your peers, including fascinating results from many departments.
Let’s talk semantic models for community financial institutions. There’s a lot of misconceptions about them. In this video, we discuss the basics of semantic models, break down myths, and share the facts on what you need to know to make an informed analytics decision for your team without getting persuaded by chatter. Exploring Common Semantic Model Questions Fielded from Credit Unions and Community Banks Gemineye’s COO Matt Jefferson and Director of Business Development Maggie Chopp answer: Whether centralized analytics teams are better than departmental analysts Why unified semantic models ensure consistency If there should ever be a cap on how many fields your semantic model should have Why the misconception of limiting number of fields is holding back your analytics potential How credit unions are successfully integrating data from multiple sources without complexity Full Transcript Alicia Disantis: Are centralized analytics teams better than departmental analysts? And Matt I will ask you that question. Matt Jefferson: I think centralized analytics teams there, they’re great for consistency and governance. Department analysts are great for kind of speed and the business contacts. I think the real win, no matter what you do in an organization, is when you really have them working with the same semantic model, the same governance principles. You know, I think at some level, you always end up with some level of decentralized, analytics. Maggie Chopp: We’ve seen a couple kind of creative configurations this year, and there’s nothing saying that they can’t work. Well, again, given the, the base semantic model like Matt mentioned, that’s kind of the key to making it all flow. But we have a number of credit unions we’ve talked to this year who have very interesting designs that work for their teams, depending on who rolls up to who. But it works nicely again, with the base model. We have a credit union that as an example that brought in data from their collections platform. A lot of that would never, ever hit the core. But it brought them a lot of value when it came to their repossessions. They ended up making a lot more money knowing which of their resellers were getting them the most margin. And again, that’s not something you’re going to find in a core provider. You’d really need a data warehouse to assess something like that. Alicia Disantis: Should there ever be a cap on how many fields your semantic model should have? Matt Jefferson: No, there should be an artificial cap. Right? Business questions evolve. You know, I always say that every financial institution is like a fingerprint, right? You know, it’s a financial institution, but it is a little bit different. Right? And the right solution, the technology solution, everything makes it very easy to adapt to your your business needs, right, to add the things you need. Do you really want to wait for someone to say like, hey, all of our clients need that particular data field before you can access it? Maggie Chopp: I’m gonna add a little caveat, which is it’s not always good to bring everything. And we want to, you know, optimize for efficiency. We want it to run quickly. Run well. But just from speaking from experience, the credit you and I worked at and all the credit runs we talked to, everybody needs something else. Like every single day. The models are great. We just need to add one more thing. And one thing that we’ve observed is that, some providers will make it really difficult to do that. You’ll have to go way back in time. There’s, a statement of work to draw up. There’s all this kind of finagling just to get one additional data field, which really, it shouldn’t be that complicated. We have a bring it all approach. And so bringing something like a single data field in is it is easy and it should be easy. And if it’s ever not, you should reassess how you’re doing it. And on top of that you know, the concept of even having placeholder fields is somewhat meaningful. We found pretty regularly talking to credit unions. They might all have a different definition of something. But if we keep placeholders in the code for different persona types or different segmentations that are credit unions already use, we don’t even have to go about adding a field, so to speak. We’re just using that credit unions logic. So no, we should never be limiting how many fields should be in a semantic model. Want More Data Analytics best practices? Check out our Data Teams page.
Microsoft Excel is where almost every financial institution’s data work begins. It is familiar, flexible, and already on every desktop. For years, it does the job. The monthly board report comes out of a spreadsheet. The branch numbers live in a spreadsheet. The incentive calculations, the loan summaries, the customer counts, all spreadsheets. And then, at some point, it stops working as well as it used to. The shift is rarely a single dramatic failure. It is a slow accumulation of friction: reports take longer, files get bigger and slower, one person becomes the only one who understands the master workbook, and small errors start having larger consequences. If that pattern feels familiar, your institution may have outgrown Excel without anyone formally deciding it was time to move on. The Signs Your Institution Has Outgrown Spreadsheets Outgrowing Excel looks different at every institution, but the warning signs are consistent. Reports that used to take an hour now take a day, because the data has to be pulled from more places and stitched together by hand. The same numbers do not match across departments, because everyone maintains their own version. A single analyst has become the keeper of a workbook so complex that nobody else can safely touch it, which means the institution has a key-person risk hiding inside a spreadsheet. In many cases it is not even an analyst doing the work but an executive, and a Chief Lending Officer who spends fourteen hours a month wrestling data inside a spreadsheet is fourteen executive-level hours not spent on strategy, lending growth, or leadership. Other signs are subtler. Decisions wait on data that is always a few weeks old by the time it is compiled. Nobody fully trusts the numbers, so meetings get spent debating whose figures are right instead of what to do about them. These are not signs of a team doing poor work. They are signs that the tool has reached the limit of what it was designed to do. Why Spreadsheets Become a Liability at Scale The core problem with relying on spreadsheets for institutional reporting is that they were never built for it. They were built for individual analysis, and they carry risks that grow as the stakes rise. Decades of academic research have reached a consistent conclusion: errors in operational spreadsheets are both common and serious, and they often go undetected until they have already affected a decision. When a spreadsheet drives a board report or a regulatory filing, an undetected error is not a small inconvenience. It is a real financial and compliance risk. Beyond errors, spreadsheets do not scale with the complexity of a modern financial institution. They cannot easily combine data from the core system, digital banking, loan origination, and third-party sources in a way that stays current. They have no built-in governance, so definitions drift and versions multiply. And they depend entirely on the people who build them, which means institutional knowledge walks out the door when those people leave. What was an asset at a small scale becomes a liability at an institutional one. What to Move Toward When Excel Is No Longer Enough Recognizing that you have outgrown Excel is the first step. The harder question for most executives is what to move toward, especially when the team has no appetite for a massive, disruptive technology project. The goal is not to abandon the kind of analysis the team does today. It is to give that analysis a stronger foundation. That foundation is a data platform that brings all the institution’s sources together automatically, maintains consistent and governed definitions, and delivers reporting that stays current without manual assembly. Instead of an analyst pulling exports and reconciling them by hand each month, the data flows in continuously and the reports update on their own. Instead of every department keeping its own version of the truth, everyone works from the same governed source. The analytical thinking the team already does remains. What changes is that the underlying work of gathering, cleaning, and reconciling data stops consuming all their time. Making the Transition Without Disruption The fear that keeps many credit unions on spreadsheets long after they have outgrown them is the fear of the transition itself. Leaders imagine a multi-year implementation, a steep learning curve, and a long period where nothing works. That fear is understandable, but it is based on an outdated picture of what moving to a modern data platform involves. The right platform is designed to integrate with the systems a credit union already runs, rather than requiring them to be replaced. Implementation is measured in weeks, not years. And because the goal is to make data easier to access, not harder, a well-designed platform reduces the learning curve rather than adding to it. The transition is not a leap into the unknown. It is a step onto firmer ground. Move Beyond Spreadsheets With Gemineye Gemineye’s Operations solution is built for credit unions and community banks that have outgrown spreadsheet-based reporting. Instead of waiting until month-end and assembling reports by hand, teams get detailed daily reporting from data that updates automatically. Manual work like branch incentive calculations, which can consume hundreds of hours a year in a spreadsheet, can be automated entirely. Because the platform unifies more than 75 data sources across core systems, digital banking, originations, and third-party vendors with consistent, governed definitions, your institution finally has one trustworthy source instead of a sprawl of workbooks. Implementation is measured in weeks, not years. If your data has outgrown Excel, see how Gemineye’s Operations solution gives your team a foundation built to scale with you.
We offer complimentary consultations – never pushy, always honest.