You can work out a social team's capacity with one sum: posts per month times hours per post, for each asset type, compared with the hours your people actually have. If the first number is above 80% of the second, you're already behind, and every SLA you promise will slip.
TLDR: List every asset type, give each one a realistic hour cost, multiply by your monthly volume, and add the unseen work (community, reporting, approvals). Compare the total with 70% of paid hours. Set SLAs per task, not per team, and only hire when the overflow is judgment work.
Count hours per asset type, not per post
"A post takes 30 minutes" hides the truth. A text post and a short video are not the same job. Time five of each type from brief to scheduled, then take the average.
| Asset type | Typical hours each | Posts per month | Hours per month |
|---|---|---|---|
| Static image or text post | 0.5 | 40 | 20 |
| Carousel | 2 | 8 | 16 |
| Short video | 3 | 8 | 24 |
| Story or repost | 0.25 | 40 | 10 |
| Production total | 70 |
These are example numbers. Swap in your own timings, because a team that reuses templates will come in far lower than one that starts from a blank page.
Add the work nobody put on the calendar
Production is about half the job. The rest is easy to forget, so give it hours too.
- Community management: 30 to 60 minutes a day per brand for replies and comments.
- Reporting: 4 to 8 hours a month per brand, less if reports are scheduled.
- Approvals and rework: 10 to 20% on top of production.
- Meetings and briefs: count them. They're real hours.
In the example above, 70 production hours plus 14 for rework, 22 for community and 6 for reporting gives 112 hours a month for one brand.
Plan against 70% of paid hours, not 100%
A full-time person has about 160 paid hours a month. Holidays, sick days, tool problems and surprise requests eat up roughly 30% of that. So plan against about 110 hours per person.
In the example, one brand needs 112 hours, which is one full person with no slack. A second brand at the same volume means two people, or a smaller scope. This is also where you see which brands cost more than they earn.
Write SLAs per task so they can be kept
A team-level SLA like "we respond fast" can't be measured. Set one number per task and one owner.
| Task | SLA | Owner |
|---|---|---|
| Brief to first draft | 2 working days | Content lead |
| Internal review | 1 working day | Brand manager |
| Client or legal approval | 24 to 72 hours | Approver |
| Comment reply | 4 working hours | Community manager |
| Urgent post request | Same day, if under 3 hours of work | Social lead |
The approval line matters most because it sits outside your team. In Mydrop's data, the median time from submission to client approval was 6 hours, and 66% of approvals landed within 24 hours. A 24-hour SLA for approvers is realistic, and it gives you a clear rule for when a late approval moves the publish date instead of your weekend.
Spot the hire signal early
Hire when three things are true at once:
- The model says you need more than 80% of available hours for two months in a row.
- The extra hours are judgment work such as strategy, community and client calls.
- Formatting, resizing and scheduling are already handled through templates and bulk tools.
If the overflow is formatting, fix the process first. Our post on doubling output without hiring covers the four changes that free up the most hours.
Make the model live in one place
A capacity sheet nobody opens is worth nothing. Keep the volume in your content calendar and give each brand its own workspace, so the monthly count comes from what's scheduled, not from memory. Mydrop's bulk creation and AI captions cut the formatting hours, which is the cheapest capacity you can buy.
Run the model this week
Time five tasks per asset type today, fill the table above with your numbers, and compare the total with 110 hours per person. Then pick the one SLA that hurts most, usually approvals, and put a number on it by Friday.
Google review
Trustpilot review
Product Hunt comment
