Strategy · Outreach · Instagram
How many follows and DMs a day the brand Instagram account can safely do, settled against what the account has already done rather than what the vendor blogs claim. Written because Roman holds 20 DMs and 30 follows as the ceiling, and the SOP says 15 and 40.
Three sets of numbers are in play, and no two agree.
| Position | DMs a day | Follows a day | Where it comes from |
|---|---|---|---|
| Roman | 20 maximum | 30 maximum | Stated 3 Nov. It is also exactly the rate being worked today, so it may describe current practice rather than a safety ceiling. |
| The SOP | 15 | 40, ceiling 50 | ig-outreach/docs/SOP-3-outreach-openers-and-analysis.md, section 3 |
| This account's own record | up to 49 | up to 44 | 347 DMs and 512 follows logged on the Instagram tab, no restriction recorded against any of them |
Note which way each number errs. Roman's DM figure is higher than the SOP's. His follow figure is lower. So this is not a case of one person being cautious and the other not.
Instagram publishes no limits, so the account's own history is the best evidence available, and it is better evidence than any blog. Reconstructed from the "DM on" and "Followed on" date stamps, because the tracker's Daily Log tab has never been filled in.
| Date | DMs | Follows |
|---|---|---|
| 22 Sep | 10 | 40 |
| 28 Sep | 11 | 21 |
| 29 Sep | 11 | 31 |
| 30 Sep | 11 | 32 |
| 1 Oct | 10 | 35 |
| 5 Oct | 21 | 40 |
| 6 Oct | 19 | 39 |
| 7 Oct | 20 | 39 |
| 8 Oct | 20 | 30 |
| 9 Oct | 19 | 44 |
| 13 Oct | 15 | 40 |
| 14 Oct | 19 | 38 |
| 15 Oct | 20 | 39 |
| 16 Oct | 28 | 7 |
| 19 Oct | 29 | 0 |
| 22 Oct | 49 | 0 |
| 3 Nov | 20 | 30 |
The 49 on 22 October was 48 private clinic rows, the bucket parked on 29 October. Nothing followed it: no restriction, no warning, no note.
The honest caveat on "nothing happened".
The Daily Log tab holds a header and one instruction row, nothing else. Its "Account Warnings" column has never been used. Every Notes and Outcome cell was searched for block, restrict, warning and paused: nine hits, all of them false positives about blockwork and block paving. So no restriction appears anywhere in the sheet, the repo, the 29 October handover or the open threads list. That is strong, but an unlogged block would look identical from here. The 12-day DM gap between 22 Oct and 3 Nov is explained by the handover and the clinics being parked, and with no log that cannot be proved either.
One row a day in Daily Log would make this settleable next time. It is the cheapest fix on this page.
Two people working the same account do not each get their own allowance.
Instagram meters the account, not the operator. Rachel at 10 DMs and 20 follows each morning, on top of Lincoln's 20 and 30 later in the day, is 30 DMs and 50 follows a day on one account. That is the number Instagram sees. It sits at the top of the estimate band for cold DMs to non-followers, above every day this account has worked except 22 October, and exactly on the SOP's follow ceiling.
The staggering itself is worth doing: spreading the same volume over two separate sessions hours apart is the opposite of the burst pattern that gets accounts restricted. The point is that the combined daily total still has to be a number someone chose on purpose, rather than the sum of two people each working to their own cap.
Nothing. Meta's spam policy is the only primary source and it gives no figures: it prohibits engaging with content "either manually or automatically, at very high frequencies", and says restrictions may come at lower frequencies "when other indicators of Spam (e.g., posting repetitive content) or signals of inauthenticity are present". Primary
One structural rule is firm and matters to the motion: a non-follower receives one text-only message and nothing further until they accept it. Follow-ups reach nobody who has not followed back, which is already why F1 and F2 are followers-only. Primary
Every number below is a vendor estimate. None of them cite Instagram, and they disagree with each other. Estimate
| Action | New account, 0 to 3 months | Established account |
|---|---|---|
| Cold DMs to non-followers | 5 to 20 a day | 30 to 40 a day, some guides claim 50 to 100 |
| Follows | 50 to 100 a day | 150 to 500 a day, around 20 an hour |
| Ramp to a stable rate | one to two weeks, raised gradually rather than jumped | |
Read these as a sanity check on our own figures, not as permission. Where they help is in confirming that 15 to 30 DMs and 30 to 50 follows is nowhere near the edge of what anyone considers risky for an account with months of history behind it.
Every source, Meta's own policy included, points at behaviour rather than volume. Two accounts sending the same number of messages get different outcomes.
The kill switch is unchanged and sits in the SOP: an action block or any warning means stop for the day, tell Rachel, and log it in Daily Log.
Roman's 30 is the one to push back on. Follows are the lower-risk action, this account has done 38 to 44 on nine separate days without consequence, and follow-backs are what move a lead out of message requests and make F1 and F2 reachable at all. Capping at 30 costs reach for no measurable safety gain.
Roman's 20 is a good number and it is above the SOP's 15. It is also today's actual rate. Raise the SOP to 20, then plus 25% a fortnight while nothing is logged, the same ramp rule the TikTok page uses. That reaches 25 in two weeks and 30 in a month, with evidence collected at each step.
No more than 20 follows in an hour, and DMs split across at least two sittings. Every source says pace matters more than the daily number, and this is the one rule neither of our docs has. Staggering the two operators' sessions does this for free.
One row per platform per day: follows, DMs, replies, and anything the app says. Without it, the next version of this argument has to be reconstructed from date stamps again, and a restriction that did happen would leave no trace.
Volume is not the lever that is underperforming. 347 DMs have produced roughly 1.8% replies, no positive ones, and no booked calls. The SOP's own threshold already says that at 150 DMs with fewer than two calls booked the approach changes rather than the list growing. More DMs a day multiplies a motion that has not converted yet, which is the argument for letting the new A/B result land before scaling anything.
Of 835 un-DMd rows that are not marked skip:
| Owner | Followed, never DMd | Not followed, not DMd | Usable |
|---|---|---|---|
| Rachel | 20 | 435 | 20. The 435 are the US roofing and HVAC rows the 29 Oct handover says to ignore. |
| Lincoln | 118 | 250 | 368 |
| Parked clinics | 12 | 0 | 0. Do not DM clinic rows. |
So the followed-but-never-messaged pool Rachel is working is 20 rows: two mornings at 10 a day. After that the choices are a reassignment from Lincoln's 118 or a UK sourcing wave. Sourcing is what unblocks scaling here, not a higher daily cap, and SOP 2 says to flag at 40 unworked rows remaining rather than waiting for the list to run dry.
Rachel works 10 DMs and 20 follows each morning, early, before Lincoln starts, so the two sessions are hours apart rather than stacked. Her focus is the rows already followed and never messaged. Ours
Possibly held until the new A/B openers have produced a readable result, since the reply rate is the open question and more volume will not answer it.
Open against this page: the combined 30 DMs and 50 follows a day needs to be a number Roman and Rachel have both agreed, not an accident of two caps added together. The hourly cap and the Daily Log rows are not in the SOP yet.
ig-outreach/docs/SOP-3-outreach-openers-and-analysis.md, daily loop and kill switch Oursig-outreach/docs/HANDOVER-2026-10-29.md, the 327 DM analysis, the parked clinics, the 435 US rows Ours