Meet Caddi in personFuture ProofSep 14–17Booth 248 · Zone 1ADVISE AIOct 20–22AI for Mid-Sized LawNov 5Legal InnovatorsNov 17–18
Discover

Caddi finds the work worth automating. And shows its evidence.

It reads the activity already in your connected tools, under the access you already have, and ranks the repeated work it found, each finding opening to the evidence behind it. Read only, always.

Request demoSign up free
Read onlyRuns in the backgroundOn the free plan
How it works

An operations analyst, not a suggestion chip.

Anyone can generate a list of things you should do. Discover works out what processes are probably running, tests each one against your systems, and only shows the ones the data supports.

  1. 01
    It reads with the access you already have

    Only the tools you connected, only what your own permissions already see. It never writes.

  2. 02
    It forms hypotheses and checks them against real data

    Caddi keeps a labelled library of the read calls it knows how to make in each tool. Discover uses it to guess what processes are running, then tests each guess against real counts and timestamps. Guesses that do not hold up are never shown.

  3. 03
    It shows its work

    Every finding that survives opens: what Caddi saw, why the work qualifies, and the arithmetic behind the estimate.

Hypotheses, tested3 kept 2 dropped
  • Morning Inbox Triage and Matter Filing1,204 messages read · 418 envelopes opened · 412 filed to the matter152 hrs / yr
  • New Matter Opening from IntakeThe same four steps in Salesforce, Clio and iManage every time117 hrs / yr
  • Trust Payment ReconciliationDeposits matched to matters on every day the ledger moved87 hrs / yr
  • Expense receipts sorted into the practice folder4 occurrences in 30 daysToo rare to rank
  • Client emails summarised before a status callNo two occurrences took the same stepsNo repeated shape
The dropped rows never reach your results. This is the whole difference between a finding and a suggestion.
The evidence card

Signal, or noise? Open one and check.

This is a finding expanded: what Caddi saw, why it thinks the work qualifies, and where the number came from. Sort by most time saved or most confident, and a shaky finding with a big number never outranks a solid one.

High confidence

Morning Inbox Triage and Matter Filing

Read each new message in Outlook, open any Docusign envelope, match it to the matter in Clio and file the executed copy to NetDocuments.

152hrs / yr5×/week · ~35 min each
Microsoft OutlookDocusignClioNetDocuments

What Caddi saw

  • 1,204messages read in Outlook in the last 30 days
  • 418Docusign envelopes opened from those messages
  • 412matters matched in Clio and filed to NetDocuments
  • 6exceptions raised for review in Teams

Why it’s worth automating

Every message already names the client and the matter. The four steps are the same each time, and the tools never talk to each other, so you carry the details between them by hand.

Per occurrence
35 min
Per week
2.9 hrs
Observed over
30 days
Build next
Rolling it out

Built to be forwarded.

The hardest part of Discover is describing it, and the second run in a firm is always whoever the first person showed. Here is what to send them.

The one-liner, for the internal post
Discover looks at the work you have already been doing in the tools you use, and tells you which of it could be running on its own. It reads under the access you already have, ranks what it finds, and shows the evidence behind every finding. Read only, always.
  1. 01
    Connect the tools you actually work in

    The picture is only as wide as the connections. Email, documents, the CRM, the practice or portfolio system, the accounting tool.

  2. 02
    Run it and keep working

    It runs in the background, and findings appear as they are confirmed.

  3. 03
    Open the top finding, then take it to the team

    Results follow the runner's access, so one run per team gives you the firm. The people who do the work know what a finding leaves out. Ask them how much of it applies, then hit Build next and Loop Studio opens with the evidence already in it.

FAQ

The questions your colleagues will ask

What does it read, and how heavy is it on our systems?

Activity in the tools you have connected, under the access you already have in each of them. It reads reports, counts and recent activity rather than walking through records one by one, so finding your processes does not turn into a spike in API usage. It does not write anything, and it does not reach a system you have not connected.

A run does not repeat on its own either. Once Caddi has read a set of tools it does not offer to read them again for 30 days unless you connect a new one, because an unchanged environment returns the same findings.

How does it estimate time saved?

From what it measured: how often the work occurred, how long each occurrence took, over the window it observed. Every finding shows those three numbers, so you can redo the arithmetic yourself.

Read it as the cost of the process, not one person’s time. Where a tool is shared, the count includes everyone who touches that work, so the figure for a shared inbox is the team’s figure. It is a way to rank findings and start a conversation, not a forecast.

Will it just tell us things we already know?

Sometimes, and that is useful. If Discover surfaces work you already automate somewhere else, it has found the real work, which is a check on everything else in the list. The findings that matter are the ones nobody had written down.

Why are my results different from my colleague’s?

Because Discover reads with your access, not the firm’s. That is why the pattern is one run per team, by someone who does that team’s work.

Is our data used to train models?

No. The contents of your connected systems are never used to train models. OpenAI and Anthropic calls run under zero-retention terms, and accounts under a Data Processing Addendum exclude their content from model development entirely.

Which plan is it on?

Discover is on the Individual plan, which is free for one user with no time limit. See pricing.

Get started

Run Discover on your own tools.

Connect what you work in, run it, and open the top finding. Or book a demo and we will run it with you.

Request demoSign up free