We’re designing Ellie to be a tool that solves problems for data teams right now. That’s why we’re building in the open, releasing roadmaps and gathering feedback while challenging ourselves.
In this blog, we’ll discuss our product roadmap for the first half of 2024 and why we’ve chosen to focus on some things.
But before we get into the roadmap, here’s a quick roundup of what we’ve just released with Ellie 5.1.

Conceptual Data Model Import

You can now do a conceptual model import to Ellie via API. The entities within the imported model can be handled in two ways:
1. You can associate your existing glossary to entities within the imported model.
2. Or you can create new entities if your current glossary doesn’t have the required entity.

‍

Why is this important?

It’s not uncommon for large companies to have dozens of data warehouses or data products, with little understanding of what’s being engineered.

What if you could unlock your data warehouse and visualize it as an ER-diagram? In quick time, you would have an understanding of:
- the processes you’re tracking and,
- what the data warehouse represents.

Update to Auto Draw

Auto draw is available when importing conceptual models or logical models and when converting logical models to conceptual models.
It was always possible for Ellie to read in database structures and “draw” a logical model based on it.

We have now improved on the previous version of “auto draw”, giving you more readable and polished models.

The Roadmap

Brand Refresh

You’ll notice a cleaner, more intuitive UI soon as we refresh the look, feel, and usability of the platform.

Personal Collection, Sub Glossaries

We’ll introduce the “Personal Collection” so that you can build models in isolation before introducing them to your team.
When you have a number of users creating models — the average Ellie customer could have hundreds of models — it’s challenging to govern and manage both the models and the glossary that’s used within.
A “Personal Collection” separates models and glossary entries such that a user can work in isolation rather than messing up the company’s collection of models and the shared glossary.

Sub Glossaries

The “Personal Collection” would be the first step in our development of “Sub Glossaries”. A Sub Glossary will enable you to separate domains (business domains, processes, user teams, etc.) and build glossary structure. This ensures only relevant users can access and edit entities that are critical to specific data processes.

Physical Modeling

We’ll expand into physical modeling, with an MVP that emphasizes the ability to get a step closer to the final data product at a faster pace. You’ll be able to create tables, SQL code, that is then pushed downstream in the pipeline.

LLMs & GenAI

Most exciting will be our foray into the use of LLMs & GenAI to create conceptual data models.
The use of Large Language Models (LLMs) lets you input text or diagrams to auto-draw a data model. In addition, you can interact with an AI chatbot to create or edit models. Subscribe to our newsletter to learn more.

‍

‍