MagikHow long before social media works?
Two clocks run at once, and almost every store reads the wrong one. A single video's fate is knowable in days — Instagram's own feature decides within 72 hours. Whether the account works takes months to answer. Judging the second on the first is the single commonest reason stores quit at week six.
The short answer
Nobody can honestly give you a number — no platform publishes one and no independent dataset measures it. What can be said is arithmetic. Reach is granted per video and the outcomes are very uneven, so judging an account means waiting for enough posts that a genuinely good one can be told apart from a lucky one. At three to five videos a week that's about two months and 25–30 posts.
A single video is different. Meta's trial reels feature shows key metrics after roughly 24 hours and decides whether to share a well-performing reel more widely based on the views it gets within the first 72 hours — that's Instagram's own operating assumption about how fast a video becomes readable. Days for a video. Months for an account.
Below: the two clocks side by side, a third clock most people start late without noticing, what question to ask in each phase, and why "90 days" is a number somebody invented.
Two clocks
These measure different things and run at completely different speeds. Confusing them produces the classic failure: six weeks of posting, a handful of quiet videos, and a conclusion that the channel doesn't work for this product.
Why the account clock is slow is worth being precise about. Distribution on these surfaces is granted per video, and the outcome distribution is heavily skewed: most posts do very little, occasional ones do a lot. Twenty posts is twenty draws on that distribution. That is genuinely too few to distinguish "our content doesn't work" from "we haven't been sampled yet" — the same reason "no views" is usually a volume problem rather than a content one.
The third clock, and it probably started late
There's a measurement clock too, and it's the one people accidentally start weeks after the others.
UTM parameters are recorded at the moment of a visit, which means tagging is not retroactive. Untagged social traffic reaches Shopify with no referrer and is recorded as "Direct", and Shopify's default attribution model — last non-direct click — excludes direct visits when assigning credit. So if you posted for six weeks before tagging your bio links, you have six weeks of posting and zero weeks of usable revenue data, and the honest position is that you cannot yet answer the question you're asking.
Tag your bio links before you start posting. It takes about twenty minutes, it costs nothing, and it's the difference between having an answer in two months and starting the measurement clock in two months. The setup is here. This is the one piece of advice in this article that is genuinely urgent — everything else rewards patience, and this rewards being early.
What to ask, and when
The useful move is to change the question as the evidence accumulates, rather than asking "is it making money" from week one and getting a meaningless answer four times in a row.
Notice that the first two rows are entirely about things you control. Whether you posted, on which surface, in which format, and whether anyone watched — those are inputs and craft. Only the bottom row depends on the market deciding it likes you. Judging yourself early on the top rows and late on the bottom one is both fairer and more useful.
Where "90 days" came from
Nowhere. It's a quarter, and quarters are how businesses plan — which is why the number recurs in marketing content without a source attached. No platform publishes a time-to-results figure, and no independent dataset measures one, because it would depend entirely on category, content quality and posting rate.
What makes two to three months a reasonable expectation is arithmetic rather than evidence. At three to five videos a week, two months is roughly 25 to 40 posts. That's the region where an outlier stops being explicable by luck. Post twice a week and the same threshold arrives in four months; post daily and it arrives in one. The timeline isn't a property of the platform — it's a property of your cadence.
| Posts per week | ~30 posts reached at | Verdict |
|---|---|---|
| 2 | ~15 weeks | Slow. Judging at 8 weeks means judging on 16 posts. |
| 3 | ~10 weeks | Workable, and sustainable for most one-person stores. |
| 5 | ~6 weeks | The point where six weeks is actually a fair test. |
| 7 | ~4 weeks | Fast, if the quality floor holds at that rate. |
This is the useful reframe. "How long until it works" is the wrong question because time isn't the variable — attempts are. Two stores starting on the same day reach a fair verdict months apart depending only on how often they post.
What doesn't take longer than you think
Two things people expect to be slow that aren't, worth knowing because they remove common excuses for waiting.
Having no followers doesn't delay you. TikTok states that neither follower count nor whether an account has had previous high-performing videos are direct factors in its recommendation system, and Instagram says the majority of what people see on Reels comes from accounts they don't follow. A brand-new account is eligible for the same distribution as an established one — the surfaces are open on day one. What takes time is the number of attempts, not building an audience first.
A single video resolves quickly. You don't need to wait a month to know whether one post worked. Meta's own trial reels flow reads metrics at roughly 24 hours and makes its sharing decision on the views a reel receives within 72 hours. If a video has done nothing after several days, it has told you what it's going to tell you — move on to the next one rather than waiting on it.
Where this leaves the work
Everything above says the same thing from different angles: the timeline is set by how many posts you make, and the failure mode is stopping before the evidence exists. Which makes the practical problem not patience but sustainability — a cadence you abandon in week five never reaches the point where it could have been judged.
That's the part Magik addresses: it generates creator-style videos from your Shopify catalogue and publishes them to TikTok, Instagram and Facebook on a schedule, so the number of attempts doesn't depend on you having a good week. It doesn't make the account clock run faster in any other sense, and it can't make a video worth watching if the underlying idea is dull.
The honest version of the pitch is narrow: automation raises the number of draws you take, and the number of draws is what the timeline is made of. Everything else on this page applies whether you use a tool or film three videos a week yourself. If you're comparing options, the comparison page covers Outfy, Xyla AI, IDEQO, Minta and Buffer alongside us.
Frequently asked
How long does social media take to work for a store?
No honest single number exists — no platform publishes one and no independent dataset measures it. Two clocks run: a video resolves in days (Instagram's trial reels read metrics at ~24h and decide within 72h), while an account takes months, because reach is per video and lumpy. In practice, two months and 25–30 posts before conclusions are worth drawing.
Why does everyone say 90 days?
Because it's a quarter, not because anything measured it. What makes two to three months reasonable is arithmetic: at three to five videos a week that's 25–40 posts, which is where an outlier stops being explicable by luck. Post twice a week and the same threshold arrives in four months.
What should month one look like?
Evidence the mechanics work, not sales. Weeks 1–2: are posts publishing, on Reels and TikTok rather than only Stories, with tagged links. Weeks 3–4: is anything being watched to the end. Expecting revenue in month one sets up a decision to quit on a measurement that couldn't have been meaningful.
How fast can I judge one video?
Days. Meta's trial reels documentation describes metrics after about 24 hours and an automatic sharing decision based on views within the first 72 hours — Instagram's own assumption about how quickly a video becomes readable. A video that's done nothing after several days has told you what it will.
When does my data actually start?
The day you tag your links, which is often later than the day you started posting. UTM parameters are recorded at the moment of the visit, so tagging isn't retroactive — and untagged social traffic lands in Shopify as "Direct", which its default attribution model excludes when assigning credit.
How do I decide whether to keep going?
Judge inputs early, outputs late. Early: are you posting at the cadence you planned, in video, on the recommendation surfaces, and is anything being watched through? Those are answerable now and within your control. Revenue is the last question, because it depends on everything upstream and on the tracking having existed throughout.
Does having no followers slow it down?
Not in the way people assume. TikTok says follower count isn't a direct ranking factor and Instagram says most Reels reach comes from non-followers, so a new account is eligible for the same distribution. What takes time is accumulating attempts — a function of posting rate, not starting audience.
Is it different for a boring product?
Probably, and honestly nobody can quantify by how much. Categories that are hard to demonstrate in video have a genuinely harder time on watch-time-ranked surfaces. The mechanics in this article don't change; what changes is your hit rate, which means more attempts before an outlier appears — a longer clock, not a broken one.
Sources
The one genuinely documented timescale here is Instagram's, and it's about a video rather than an account. Everything about the account clock is arithmetic derived from how per-video distribution works, and is labelled as such.
- Meta Newsroom — Test content with non-followers using trial reels, 10 December 2024. Source of the roughly 24-hour metrics window and the automatic sharing decision "based on the views it receives within the first 72 hours", which is the only published platform statement we found about how quickly a video's performance becomes readable.
- TikTok Newsroom — How TikTok recommends videos #ForYou. Source of the statement that neither follower count nor previous high-performing videos are direct factors in the recommendation system.
- Instagram — Instagram Ranking Explained, 31 May 2023. Source of the statement that the majority of Reels content comes from accounts you don't follow.
- Shopify Help Center — Measuring marketing performance. Source of the last non-direct click default and its exclusion of direct visits, which is why untagged traffic can't be credited retroactively.
- Not cited, because it doesn't exist: a time-to-results benchmark. No platform publishes one, and the "90 days" figure in general circulation has no traceable source. The account-clock thresholds in this article — ~25–30 posts, two months at 3–5 a week — are arithmetic from the per-video, high-variance nature of distribution, presented as reasoning rather than measurement. The week-six mark in the first diagram is illustrative of a common pattern, not a measured churn statistic.
Keep reading
How often should a store post?
Cadence sets the timeline — the same threshold arrives in one month or four depending on it.
Is social actually making you sales?
The measurement clock, and why it's the one thing here that rewards being early rather than patient.
Why do my videos get no views?
If the answer is "nothing is happening", this separates too-few-attempts from something structural.
Do you need followers to sell?
Why a new account isn't starting from behind — the surfaces are open on day one.