I’m an Emma Ultimate user managing household finances across several accounts. We now use an agent to audit Emma’s Google Sheets export, identify classification errors and model budgets, but every correction and taxonomy change must still be entered manually.
A scoped personal API—or even taxonomy import/export—would make Emma considerably more useful without requiring open-ended banking access. Our immediate use cases are:
• list/create/update categories and subcategories;
• list/create/update tags and category budgets;
• bulk reclassify transactions and manage “apply to similar” rules;
• incremental read access to transactions and balances using stable IDs;
• webhooks for changed transactions;
• dry-run, idempotency, audit history and revocable read/write OAuth scopes.
A smaller first release could simply support JSON/CSV import/export for categories, subcategories, tags, budgets and classification rules. That would remove a large manual setup burden while keeping financial-account connectivity entirely inside Emma.
Is any supported API, partner programme or structured taxonomy import planned? I would be happy to test an early version and share a concrete agentic workflow.
Wow that’s quite a laundry list. Half of those (AI-written) ask are not implemented in the backend in the first place.
My most immediate question would be though, why would Emma as a business invest in any of this? When I propose something to a business (or anyone really) I have to think about the upside for them too. Theit monetisation model relies on the app being used as a frontend. Productising their engine just for a few pro users won’t justify the investment I am afraid..
Thanks, Lucian. That is a fair challenge, and I appreciate the honest reply.
I did use an agent to help structure the post, so apologies if it came across as an unrealistic laundry list. The intent and the use case are genuinely mine. I am actively trying to make Emma the main source of truth for my household finances and to use a personal agent for analysis, forecasting and routine financial review.
I agree that building and maintaining a full API for a small number of advanced users may not make commercial sense. I was thinking less about charging for API traffic and more about whether this could make Pro or Ultimate subscriptions more valuable and harder to replace. Emma could remain the main app where users connect accounts and review or correct data, while agents perform analysis that Emma would not necessarily want to build itself.
Perhaps a more realistic first step would simply be a stable, documented, read only version of the data already available through Live Export, with reliable transaction IDs, categories, tags and pending or posted status. That would cover much of my practical use case without requiring Emma to productise its whole backend.
In any case, thank you for taking the time to respond. Your point about considering the benefit to Emma as a business is well made.
I strongly agree with Timur here. I think an API (or at least a CLI) would be extremely useful, so I can ask agents to update my transactions and manage my finances for me.
I think it could even be the first reason I could think so far to consider upgrading my membership to Ultra (if it was exclusive to that tier).
I spend a lot of time manually updating my transactions to ensure I categorize everything properly and know where my money is going.
I’ve provided feedback about improving the experience of bulk editing the transactions, but, TBH, being able to use agents is a lot more efficient, bc you can connect them to data sources that will provide the Notes for the transactions, e.g.
Amazon order emails: the agent can add a note for what you bought
Flight tickets: the agent can add a note for what trip it is.
Digital purchases: the agent can read the receipts from the email and add a note saying what it was for.
All this to say that providing a way for agents to interact with my Emma would make my usage significantly more efficient and productive, and I think that adds a lot of value to the product, so I would be willing to pay a bit more.