Friday afternoon, and half the team is rebuilding the same status deck they built last Friday. The data came from three client systems, got pasted into a spreadsheet, got summarized by hand. Every hour spent rebuilding it from memory is an hour not billed and not sold — and it's the same hour, every week.
Describe the work in plain English — "every morning, tell me which engagements are overdue a client update, which are blocked, and what utilization looked like last week" — and InTouch AI builds the workflow, runs it on schedule, and delivers it. No scripts. No per-report bill. And client data stays on your server.
InTouch AI lives on your own computer, not someone else's cloud. The data you pull from a client's systems to build a deliverable — often under an NDA — never leaves your office, and nobody else can get to it. For a firm whose whole product is trust with sensitive information, that's not a nice-to-have — it's the contract.
Everything below is a real, ready-to-run automation you install with a click and point at your own engagements — your clients, your rate card, your utilization target. You adjust it and check it first, and nothing reaches a client without your review. The deliverable has your name on it.
The value is the thinking. The slippage is in the pieces — the section that never arrived, the appendix owned by someone on another job, the pack that was missing one exhibit at 4pm on the day it shipped. On a fixed-fee engagement nobody notices until it is late.
At six people there is no system of record for status — it is in Slack threads, a calendar, and whoever ran the call. So the Friday update gets written from memory, and the engagement that has gone quiet for three weeks is the one nobody remembers to mention.
Clients want to know where things stand. Producing that update — gathering progress, writing it up, sending it — is a recurring tax on every engagement, done by the most expensive people in the room.
Who's over-allocated, who has room, which engagement is bleeding hours — you find out at month-end, when it's too late to adjust. A weekly read would let you steer instead of explain.
Describe the question, get a structured answer. Paste the input, read the result. Point it at your AI provider — or a local model, so nothing leaves the office at all.
Turns a raw transcript into structured notes — action items with owners, decisions made, open questions, key topics. Built for what Zoom, Teams and OBS actually produce. The discovery call written up before you have left the car park.
Pulls the clauses that decide whether an engagement is worth taking — term, renewal, liability, termination — out of a client's paper. A first read that tells you where to look, not a substitute for your own or your counsel's.
Scheduled or trigger-driven. Each reads a source you already have — a Sheet your firm keeps, a folder your work lands in — does the work, and delivers to the channel you live in. It runs whether Friday is calm or slammed.
Sorts every active engagement each morning into what is blocked, what is overdue a client update, and what is up to date — with the phase, the next milestone and how far away it is. The drafts come to the engagement lead. Nothing is sent to a client on its own.
Hours, billable hours, utilization against target and realized revenue per person for the week, plus who is under target and who is over capacity. No AI anywhere in it — utilization is a number a partner acts on, so it is arithmetic or nothing.
Prices the scope from your rate card in code, then writes the SOW around the computed figures and your discovery notes. The model never prices anything. A role with no rate on the card is flagged by name, not guessed at.
Checks each engagement's folder against what the client is owed and tells you what is missing, who owns it and what is late — broken out per owner, so four people chasing four pieces get four lists. A presence check; it does not open the files.
Grab the free Personal edition of InTouch AI and run it on the firm's server or your laptop. Browse the InTouch Hub, pick the skills and workflows you want, click Install. No scripts to write, no engineer to hire.
Each workflow ships with placeholders — data sources, engagement list, report templates, recipients. Read the README, fill them in, point it at your systems. Verify a run produces the deliverable you'd sign before you schedule it.
Attach a weekly or per-milestone schedule and walk away. InTouch AI runs it, delivers the digest or pack, and logs every run. When a client's export format shifts or a token expires, InTouch AI reads the failure, fixes what InTouch AI can, and tells you why. As a workflow earns trust, dial it toward fully deterministic: same result every time, zero per-run AI cost, fully audited.
Straight talk. No surprises after you install:
Personal edition is free. Install it on the firm’s computer, point it at the sheets you already keep, and let it surface the engagements overdue an update, the deliverables missing a piece, and what utilization actually was last week. Every draft comes to you, not to a client. And because it all runs on your own computer, client data — often under NDA — never leaves your office.
Every one installs from the hub and runs on your own server. Browse all over a thousand connectors, or describe what you need and the assistant builds it.
All three time trackers are hand-built, not generated from a spec — so Utilization Tracker reads your real hours instead of a re-keyed spreadsheet. Each one paginates properly and tells you when a result is incomplete, because a utilization figure built from a truncated first page under-reports and still looks reasonable. Not on any of them? The Sheet version of the same workflow does the job.