Umfeld is a graph-based CRM, not an address book
Google Contacts stores how to reach someone. It does not store how you are connected, who introduced you, or who else in your circle already knows them. That is what a graph captures, and a list structurally cannot.
A contact list answers exactly one question: how do I reach this person? Each row stands alone: name, number, email, done. That works for a phone book. It is the wrong shape for a network of people.
The limit shows up the moment you ask who knows whom, who actually introduced you to someone, or which three people in your circle have quietly known each other for years. A list has nodes, not edges. Every relationship between two people ends up crammed into a free-text note, unstructured, and two years later even you cannot find it again.
Umfeld flips that around. A relationship is not a sentence in a notes field. It is its own object in the database.
A list has no edges.
Relationships as objects, not notes
In Umfeld, knows, partner, ex-partner, colleague, mentor, related to, and introduced by are not labels typed into a notes box. They are edges between two contacts, each with its own fields: since when, until when, how close, and how you met, through which person, on Tinder, on Instagram, at a conference in Lisbon.
The distinction sounds pedantic until you try to use it. Who introduced me to Klara is a memory exercise against a list. Against a graph it is a query that returns in milliseconds.
It is also the difference from a Notion database with relations: two entries link to each other, but the link itself stays empty, no closeness, no time range, no context on how you met. More on that in the comparison with Notion.
Who knows whom
The real payoff shows up once two edges meet. Does Mira know someone who also knows Lukas? How many hops separate you from a specific person? These are not exotic questions. They come up constantly: networking, figuring out who you can invite to something without it getting awkward, tracing how an introduction actually reached you. A list cannot answer them, structurally. A graph can, because the connection itself is stored, not just the two people at either end.
Example
Lukas knows Mira (introduced by: Jonas, 2019)
Mira works with Sofia
Sofia knows Lukas (met at: a conference in Lisbon)
Question: who connects Lukas and Sofia?
Answer: Mira, since 2019, through Jonas.
The timeline is a derived view, not a form
A contact profile in Umfeld is not a form you dutifully fill in. It is a view over the graph: everything that has happened with this person, in one place. Every logged meeting, every life event, every task, every gift hangs off the same contact as a node, and together they add up to a timeline that writes itself.
You enter the individual facts. The story assembles itself when you need it, whether that is a meetings timeline or the answer to a question you would otherwise have had to ask yourself.
The graph is what makes the AI good
A language model that only sees a contact list can look up names. That is the ceiling. A model that sees the graph can answer questions you would otherwise have to sit down and work out yourself: who in my network knows someone at this company? Who have I not spoken to in months who actually mattered to me? That is exactly why Umfeld connects to your AI over MCP. More on that under connect your AI.
FAQ
Do I need a graph if I only track twenty contacts?
Does the graph replace Google Contacts?
Can I actually see the graph?
See your own graph.
Connect your Google contacts, and Umfeld builds the graph from what is already there.