Harlow & Finch
INR 2,10,00040%
For agencies
You work in your client’s account. The bill comes to you.
Do this week for Amber Home.
- How they are doing
- This week against their normal
- Their best posts
- Plan their week3
- Write the drafts
- Put them in the calendar
Fieldnote, your agencywho pays
Amber Home, your clientwhose account
Turn either one off.
- jobs that ran
- 0 of 6
- billed to Fieldnote, your agency
- nothing
If either subscription stops, you can still look at everything. It tells you what it would have done, and shows you the numbers that answer the same question.
Both accounts are paying, so all six jobs run, and they come off your agency's plan, not your client's.
The stricter of the two wins. Otherwise the assistant would be a way into an account that is supposed to be shut.
The narrowest permission wins, and it is worked out once. Who you are is worked out by the same code every other part of the product uses, because a second copy would be a second set of bugs.
Your whole book, at once
Which client needs you today.
Each client is drawn at the size of what they pay you every month, so a quiet fortnight on your biggest client does not look the same as one on your smallest.
Meridian Skincare
INR 1,40,00027%
Nine Yards
INR 65,00012%
Amber Home
INR 48,0009%
Pell & Sons
INR 36,0007%
Roam Coffee
INR 22,5004%
No client has an overdue invoice, an expiring contract, a deadline in the next three days, or a quiet fortnight. Nothing needs attention right now.
- need you today
- 0 of 7
- open first
- nobody
- money owed
- agreements running out
- deadlines arriving
- gone quiet
Four things, and only the four that cost you money or a client. A serious one counts three times as much as a small one.
Vestry Gin: none recorded. No monthly fee for them, so they are not drawn to scale. A missing fee just means nobody has typed one in yet.
A client with nothing wrong does not appear here.
Profit per client is not here. Nothing in this product records the hours you spend on a client, so without hours every client would look 100% profitable, which is useless.
The triage screen and the assistant run the same single calculation rather than one each, so they cannot disagree about whose invoice is overdue.
No absolute numbers anywhere
Compare each client against your other clients.
A client with 40,000 followers is not underperforming for being smaller than one with 400,000. So there is no fixed target here. The comparison is against your other clients, and it moves as they do.
Followers
the middle50,0503 above · 3 below
1,48,000+97,950
96,400+46,350
61,200+11,150
38,900−11,150
24,500−25,550
11,300−38,750
Pell & Sons
not in yetnot compared
Pell & Sons is not compared, because their follower count has not come through yet.
With only 6 clients to compare, this is a rough guide rather than a verdict.
Accounts with no engagement figure yet are left out rather than counted as zero. Most accounts start with no figure, so this happens often.
The only comparison that matters is against your own clients.
A median and a percentile rank, taken only against the agency's own clients. No absolute thresholds exist anywhere in this tool.
Checked by running it
Your clients cannot see each other.
We cannot check ninety tools by reading them, so we test them. Every client account is given a made-up word that appears nowhere else. Then we run every tool as eight different people and search every answer for those words. If one client's word turns up in another client's answer, the test fails.
- ZQCREATORONLYa creator: not found in any answer
- ZQBRANDONLYa brand: not found in any answer
- ZQAGENCYONLYthe agency: not found in any answer
- ZQCLIENTONLYa client of theirs: not found in any answer
- ZQCLIENTTWOONLYa second client: not found in any answer
- ZQOUTSIDERONLYan unmanaged workspace: not found in any answer
- ZQCREATORMEMBERONLYthe creator's team member: not found in any answer
- ZQAGENCYMEMBERONLYthe agency's team member: not found in any answer
- control · 15 readsThe control found nothing, so the test itself is broken and this run does not count.
Only the words belonging to a creator turned up in the answers. The other 7 accounts had none of theirs.
A tool that quietly returns nothing looks exactly like a tool that correctly refused, so the test also forces every tool to fail loudly.
The control account is the one where the word is supposed to appear. If the control finds nothing, the test itself is broken and the run does not count.
Bugs that leaked across accounts have shipped in this product before. That is why this is a test and not a promise.
An automated test fills a separate copy of the database with two agencies' clients, runs every role against it, and then searches every answer for the other agency's data.
Ad spend across the book
One number nobody here can add up.
Six clients, two currencies, thirty days of media. Watch what happens where the total goes.
- 01Amber HomeINR₹1,84,500
- 02Roam CoffeeUSD$3,120
- 03Nine YardsINR₹96,200
- 04Harlow & FinchUSD$8,940
- 05Slow MorningsINR₹42,800
- 06Kettle & CoUSD$1,475
one total₹3,37,035
- INR · 3 clients₹3,23,500
- USD · 3 clients$13,535
Refusing to add INR and USD: the result would be a number that looks correct and means nothing
There is no exchange rate in this product, and inventing one would be worse than refusing.
Grouped by currency, and both groups are real numbers.
Adding two currencies together throws rather than guessing a rate; the roll-up groups by currency instead. The roll-up is GET /api/ads/agency-rollup, read by AgencyAdSpendCard.
Run your book from one place.
Connect a client and the first answer is waiting before you have finished looking around. Nothing crosses between them, ever.
Cancel any time. Every dashboard and report is free to open.