In my retail store I paid the social media manager by a formula: a target monthly pay, 13 weighted metrics, and three performance levels. It lived in an Excel sheet. It gave both of us clarity from the first month, and it also taught me what a pay formula can and cannot do.
This page is the whole thing: the real metric thresholds and weights, what happened in practice, what I would change now, and a cleaner version you can adapt to your own business. I ran that business for 15 years and built pay rules for every role in it. The SMM formula below is the one that later became the prototype for PayByResult.
Three ways to pay a social media manager
Fixed salary. Simple and predictable. The weak point is that "good work" stays in the owner's head, so every conversation about results turns into an exchange of opinions.
Fixed salary plus a bonus for one number. Followers, reach, or sales. Easy to explain and easy to optimize for. One number rarely describes the whole role, and the person will deliver exactly that number.
Target monthly pay with weighted metrics and three performance levels. More setup, but both sides see what "good" means before the month starts, and pay follows results in both directions. This is the option I used, so the rest of the page is about it.
The real formula: 13 metrics, three levels
Context: a retail store, Instagram as the main channel, with Facebook, YouTube, TikTok, Telegram and Google Business alongside. The metrics, weights and thresholds below are the real ones. The money is shown at a target monthly pay of $1,500 to keep the example readable: every metric at minimum pays $1,050 (70%), every metric at maximum pays $1,950 (130%). Each metric had its own three thresholds and its own share of the money. To earn that share at a given level, the metric had to reach the threshold for that level.
On top of the formula there was a sales bonus: 5% of sales that came through Instagram direct messages, minus returns.
| Metric (per month) | Weight | Minimum · 70% | Target · 100% | Maximum · 130% | Payout at target |
|---|---|---|---|---|---|
| Organic reach (accounts reached) | 12% | 14,000 | 20,000 | 26,000 | $180 |
| Organic engagement (interactions) | 12% | 1,050 | 1,500 | 1,950 | $180 |
| Feed posts | 12% | 8 | 12 | 16 | $180 |
| Reels | 12% | 6 | 8 | 10 | $180 |
| Stories | 12% | 147 | 210 | 273 | $180 |
| Customer reviews | 10% | 7 | 10 | 13 | $150 |
| Customer tags, mentions, reposts | 5% | 7 | 10 | 13 | $75 |
| Telegram posts | 5% | 6 | 8 | 10 | $75 |
| DM inquiries | 4% | 21 | 30 | 39 | $60 |
| Facebook posts with a product link | 4% | 14 | 20 | 26 | $60 |
| YouTube videos | 4% | 6 | 8 | 10 | $60 |
| TikTok videos | 4% | 6 | 8 | 10 | $60 |
| Google Business posts | 4% | 8 | 12 | 16 | $60 |
| Total | 100% | $1,500 |
Thresholds are rounded to whole units. Reach and engagement are monthly counts (accounts reached and interactions), not money. At minimum a metric pays 70% of its target share, at maximum 130%.
What actually happened
The first month was the best argument for the formula. The manager could open the sheet, see what was expected, and see whether the month was on track. Pay conversations stopped being about impressions and started being about numbers we had both agreed on in advance.
Then the pattern showed up. Within the first months, the output metrics (posts, reels, stories) quickly reached or came close to their maximum. Reach and engagement did not move the same way. Later, the person who worked under this system described it in one word: a race. Content was being produced to hit the count.
I can't blame her for that. Look at the table again: eight of the thirteen metrics, carrying 57% of the weight, were about volume the manager fully controlled. Reach, engagement, reviews and customer tags, the things I actually wanted, carried 39% and depended on the platform and on customers. DM inquiries, the only metric close to a business result, carried 4%. The formula paid mostly for the easy part.
The formula was not useless. It showed me exactly where my own judgment had been wrong.
What I would change now
The mistake was mixing three different kinds of things into one list of metrics. They need to be separated before deciding what affects pay.
- Work the person controls directly. Posts, stories, response time. Keep it in the formula as a floor, with a small share of the money. It tells the person what the minimum pace is, and it should not be where the upside lives.
- Quality the person strongly influences. Saves, comments, DM inquiries, customer tags, reviews. This is where most of the weight belongs, because this is what "good content" means for a business.
- Platform outcomes the person can influence but never fully control. Reach, follower growth. Algorithm changes, ad budget and market swings move these numbers more than one person does. Low weight, or a separate bonus rather than a core metric.
Two more changes. Thirteen metrics is a spreadsheet, not a priority list; five or six is enough for one role. And one business-result metric, such as sales through DMs, deserves real weight inside the formula rather than living only as a side bonus.
There is also a simpler structure I did not use: a stable rate for the agreed content plan, plus a bonus for content that clearly outperforms. It avoids the race. It also gives less month-to-month signal. Both are legitimate. Choose based on whether you need the person to optimize toward targets or to hold a steady pace with occasional upside.
A cleaner version you can copy
This is what a rebuilt formula for the same role looks like, in the structure PayByResult generates. Same role, same $1,500 target pay, same three levels. What changes is the structure: six metrics instead of thirteen, and most of the weight moved to results.
| Metric (per month) | Weight | Minimum · 70% | Target · 100% | Maximum · 130% | Payout at target |
|---|---|---|---|---|---|
| Average engagement rate | 25% | 2.5% | 4% | 6% | $375 |
| Revenue attributed to social media | 25% | $2,000 | $3,500 | $5,500 | $375 |
| Follower growth | 20% | 3% | 5% | 8% | $300 |
| Content posts | 15% | 45 | 60 | 80 | $225 |
| DM response time and customer reviews | 15% | two metrics, 15% combined | $225 | ||
| Monthly payout | 100% | $1,050 | $1,500 | $1,950 | |
Half of the money sits on results the person influences directly: engagement and attributed revenue. Content volume is a floor at 15%.
One thing I would still adjust, and it is a good example of why the owner has to make the final call: follower growth carries 20% here. If your growth depends mostly on ad spend, cut it to 10% and move that weight to engagement or DM inquiries. The formula gives you the structure. You know your business.
Four mistakes I would avoid
- Paying mostly for volume. You will get volume. My 57% on output metrics is the clearest example.
- Tying most of the money to numbers the algorithm decides. Reach and followers can drop for a month with no change in the quality of the work.
- Changing the rules mid-month. New priorities get discussed before the month starts. Adjusting targets after the results are known turns a formula into moving goalposts.
- Too many metrics. When everything counts, nothing is a priority. Fewer metrics with honest weights do more.
How to build one for your own social media manager
- Set the target monthly pay. What you would pay for a solid month with every target met. That number anchors everything else.
- List what the person can actually influence. Content plan, replies, DM inquiries, reviews, saves, attributed sales. Leave out what they cannot move.
- Pick five or six metrics and give them weights that add up to 100%. Put most of the weight on quality and business results, a small share on volume.
- Set three levels for each metric. Minimum, target, maximum. Use the last three to six months of real data so the targets are achievable and the maximum is a stretch.
- Agree it before the month starts. Then log actual results at the end of the month and calculate the payout. Adjust the formula only between months.
Before making part of pay variable, check local employment and minimum-wage rules. In many countries the fixed part of pay must stay at or above the legal minimum, and variable pay needs to be documented in the employment agreement.
PayByResult does steps two to four as a first draft: describe the role, get metrics, weights and three levels, then change anything that does not match your business. AI gives the structure. The owner brings the judgment.
Frequently asked questions
How much of a social media manager's pay should be variable?
In my own formula the floor was 70% of target pay if every minimum was reached, and the ceiling was 130%. A variable part of 20 to 30% of target pay is enough to create a real signal. Check local employment and minimum-wage rules: in many countries the fixed part must stay at or above the legal minimum.
Should I pay a social media manager per post?
Only if posting is the whole job. Per-post pay buys volume, and volume is the easiest thing for a social media manager to produce. If you also want reach, inquiries or sales, those need their own weight in the formula.
Should follower growth affect a social media manager's pay?
With a small weight, or as a separate bonus. Follower growth depends on the platform's algorithm, ad budget and market changes more than on one person's work. Engagement, DM inquiries, customer tags and reviews are closer to what the person actually influences.
Does the same pay logic work for a freelance social media manager?
Yes, as a retainer structure: a fixed monthly rate for the agreed content plan, plus one or two result metrics with a bonus. The point is the same: agree in advance what good work means and how it changes the payout.