The most expensive line item in a service business is not payroll. It is waiting.
I once watched a $40,000 project sit untouched for eleven days because the client could not find their own logo file. Nobody was on vacation. Nobody was being difficult. The work was parked somewhere only one person could unpark it, and that person was the founder.
That is the whole problem. Not laziness. Not bad clients. A founder operations system that runs on your memory.
Missing assets are not a client problem, they are a systems problem
Logo files, copy, platform access, and approvals. Four things. They stall almost every project I have ever seen, and they stall it in the same order every time.
Client onboarding delays feel personal. They are not. They are structural. If the only way to get a client's brand assets is to send a text that says "hey, can you send me the logo," then the project now depends on your mood, your memory, and their inbox.
When I ran NYIM Training in New York City for 18 years, we trained more than 90,000 people. The classes started on time because the material never depended on one student showing up. Service work should be built the same way. The project should not depend on whether you remembered to follow up on a Tuesday.
The first fix is boring. Put every project in one database with a real status field.
Airtable, Notion, or a HubSpot pipeline will all do it. The status is never "in progress." It is "waiting on client," "waiting on us," or "approved." Three states, one owner each.
Once you can actually see how many projects are parked in "waiting on client," the number will annoy you. Mine did.
One front door beats five group chats
Most founders collect assets through email, text, WhatsApp, Slack, and a phone call nobody wrote down. Then they wonder why the same question comes back three times.
Build one front door. A single intake form for every new client: Tally, Typeform, or a form inside Airtable. It asks for the same twelve things every time, in the same order, and it does not care what mood you are in.
For the asset side, I like Content Snare because it chases the client for you and shows them exactly what is still missing. If you would rather build it yourself, a Notion portal with a checklist works fine. So does a shared Google Drive folder with a locked structure that clients cannot rearrange.
Access is the step people skip. Do not collect passwords in email. Use 1Password or Keeper and send a guest link. It takes four minutes and it saves you the two hours you will spend hunting for their ad account login in March.
The rule that makes all of this work: nothing enters production until the checklist is green. No green checklist, no kickoff call. That one sentence removed most of the friction from my own businesses.
If you want the deeper walkthrough of how these systems get built, I recorded the whole thing at https://workshop.mastermindshq.business.
Automate the follow-up so it stops being a personality trait
Here is where AI automation for service businesses actually earns its keep. Not in writing your proposals. In chasing.
The workflow is simple. A project hits "waiting on client." That status change triggers a sequence in Zapier, Make, or n8n. Day one, a friendly note. Day three, the same request with no emotion attached. Day seven, a calendar link and a phone call. Day fourteen, a note to your own team that the project is at risk.
Two things make this work.
First, every ask has a date. "Can you send the headshots" becomes "I need the headshots by Thursday." A request without a date is a wish.
Second, every ask has a default. If the client does not respond by Thursday, we use the stock photos we already licensed, or we push the launch one week and the fee stands. You decide that in the contract, not in a frustrated Slack message on a Friday afternoon.
For the message itself, I write one good version and let Claude or ChatGPT rewrite it five ways so it does not sound like a robot. Same ask, slightly different words, sent by a system that never takes it personally.
That is how you stop chasing clients for content without becoming the person who nags.
Approvals should be one click, not a hunt through email
Approval is the quiet killer. The work is done, the client likes it, and it sits in a thread for nine days while everyone waits for a "looks good."
Give them a button. Frame.io for video review. PandaDoc or Dropbox Sign for scope changes. An Airtable interface or a Notion page with a single Approve button for everything else. If the client has to reply to an email to say yes, you have not built client approval workflow automation. You have built a nicer inbox.
Add the clause that saves you. Silence for five business days counts as approval as delivered, and changes after that become a new scope. Say it once, kindly, at the start, and almost nobody argues.
I also record a two minute Loom walking through exactly what I need them to look at. Approval rates go up when people know what they are approving.
The part I will never automate
Everything above handles the boring parts. It does not handle the relationship.
So I keep one human touch per week per client, and it is not about the project. "How did the move go?" "How is your mom doing?" Real questions, sent by me, not by a sequence. Because of that, clients answer the automated emails too, because they know a person is behind them.
What is becoming clearer to me is that the people who scale are not the ones with the most discipline. They are the ones who built a system for the forgettable things so they could spend their attention on the people they love and admire.
The system does the chasing. You do the caring.
If you are weighing a course against a room full of operators who are actually running these systems, I wrote about the difference here: https://www.mastermindshq.business/mastermindvsl.
If this resonates and you want to build this kind of infrastructure in your own business, the Mastermind is where we do the work live. You can learn more at mastermindshq.business.