"I'll send that over tomorrow."
"We'll revise the proposal."
"Let me check with the developer and get back to you."
"Send me the document and I'll approve it."None of these feels big enough for a ticket. But forget one and the client notices, often before you do. A missed small promise can hurt trust more than a late big task, because it feels personal. "They said tomorrow. It's been a week."
Why promises slip
Project work comes with structure. Features get tickets, designs get deadlines. A promise is usually one sentence at the end of a call. Everyone hangs up, goes back to work, and the commitment ends up in meeting notes, an email thread, or nowhere.It's not a discipline problem. The promise just never became a next action.Listen for the trigger phrases
Once you start noticing them, you'll hear them everywhere, from both sides of the call:- "I'll send..." / "We'll get back to you..."
- "Let me check and..." / "I'll look into..."
- "Once you send..., we can..."
- "Can you share... by Friday?"
- "We'll decide by..."
- "I'll introduce you to..."
Write it down before the next call starts
A note that says "send revised estimate" isn't enough. Make it:- Next action: Send revised estimate to Northwind
- Due: Friday
- Reason: Promised on Wednesday's call
Promises go both ways
Every client relationship has two lists of commitments.You owe the clientThe client owes you
Proposal or estimateApproval of design or scope
Progress updateCredentials and access
Answer to a questionFeedback on a draft
Document or fileAssets and content
Meeting scheduledPayment of an invoice
Waiting isn't the same as forgotten
Monday: you ask for approval. The client says "I'll take a look." You tick off your task, because your part is done.That's exactly how approvals vanish for three weeks. Instead, write:- Waiting for: homepage approval
- Since: Monday
- Next check: Thursday
Not every promise belongs in the project tool
The natural reflex is to dump everything into ClickUp, Asana or Jira. It works for a while. Then your delivery board fills up with "follow up with Sarah," "ask John for approval," "check if Mike replied," and the real work gets harder to see.Those aren't deliverables. They're the conversation around the work. Many small agencies find it easier to keep them apart: the project tool answers "what do we need to build?", and something else answers "what does this client need from us next?" More on that split in CRM vs project management for agencies.Keep the reason with the reminder
"Contact Sarah" is better than nothing. But you still have to work out why. Compare:Waiting six days for approval of homepage version 3.Now the email almost writes itself:
Hi Sarah, just checking whether you've had a chance to look at the latest homepage. If it works for you, we can start building next week. Happy to talk through anything first.
When you're the one running late
It will happen. What matters is how you handle it. Tell them early, give a new date, keep it short:Hi Alex, the revised estimate is going to land Monday instead of today. Wanted to let you know now rather than leave you waiting. You'll have it by noon Monday.People are much more forgiving of a heads-up than of silence followed by "sorry, this slipped."
The promises you didn't know you made
The dangerous ones aren't "I'll send it Friday." They're the half-sentences the client heard as a commitment:- "I'll look into it." The client hears: you'll come back with an answer.
- "That should be possible." The client hears: it's included.
- "Let's revisit that later." The client hears: you'll bring it up again.
- "We can probably do that next sprint." The client hears: next sprint.
- Translate on the spot. "I'll look into it" becomes "I'll check with the developer and tell you by Thursday whether it's possible." If you can't give a date, say so: "I don't know yet, I'll tell you by Thursday when I'll know."
- Close the ones you won't keep. If you said "let's revisit that later" and you're not going to, retract it explicitly: "On the dark mode idea we talked about: we're not going to include it in this phase. If it's important, let's scope it separately." Retracting a soft promise costs one awkward line. Letting it rot costs trust when the client brings it up in month three.
Pad the date, not the scope
When you name a date, pick the one you'd hit on a bad week, not a good one. Our instinct is reliably optimistic. In a well-known study of the "planning fallacy," students estimated how long their thesis would take: on average 34 days, and 49 days if "everything went as poorly as it possibly could." The average actual time was about 55 days, longer than even their worst case (Buehler, Griffin and Ross, 1994). Your "I'll send it tomorrow" is subject to the same bias. "Thursday" when Wednesday is realistic. Clients remember whether you hit the date far more than they remember how early it was. And when a promise is big, split it: "I'll send the first two sections Tuesday and the rest Friday" gives you two small promises you can keep instead of one large one you might miss.Five minutes each morning
No need for a big weekly review. Just scan:- Promises you made that are due today or already late
- Client questions still waiting for your reply
- Things you're waiting on, and for how long
- Anything that's been waiting long enough to deserve a nudge
Try this now
Look at your current clients and answer, without opening anything:- What did you last promise each of them?
- What are you waiting for from them?
- When will you talk to them next?
- Which one is overdue right now?
Common questions
Should I log promises in my project management tool?
Put the deliverables there. The reminders, check-ins and approvals around them work better somewhere built for client follow-ups, so your board stays about the work.How detailed should a promise record be?
One line. Who, what, when, and where it came from. If it takes more than a minute to write, you'll skip it.What about promises made in Slack or email?
Same rule. When you read the commitment, create the next action right then. "I'll do it later" is how it got lost the first time.Where promises live in AwiFin

Sources
- R. Buehler, D. Griffin and M. Ross, Exploring the "planning fallacy": Why people underestimate their task completion times, Journal of Personality and Social Psychology, 1994.