How long to run a Meesho ad campaign.
Give a campaign at least 7 days before you judge it, never kill it in a day or two, set No end date on proven winners, and review weekly. Here is why the learning window decides everything.
Run a Meesho ad campaign for at least 7 days before you judge it, and never kill it in the first day or two. The ad system needs a learning window to settle, so early results are noise. Set No end date on proven winners and review every campaign weekly.
- Give every campaign at least 7 days before judging; the early days are a noisy learning window.
- Do not kill a campaign in 1 to 2 days, a slow start often steadies by day seven.
- Use No end date for validated, profitable winners so they keep earning continuously.
- Use a fixed end date for tests, promotions and seasonal pushes that should close themselves.
- Review weekly, and reconcile the numbers first so the verdict rests on real profit.
Learn, judge, then scale or pause
Every campaign moves through the same stages. The mistake is trying to skip from launch straight to a verdict, before the learning window has even closed.
The hardest part of running Meesho ads is not setting them up. It is having the patience to leave them alone long enough to learn whether they work.
Why a day or two tells you nothing
When a new campaign launches, the ad system does not yet know who your catalog appeals to or which placements will convert. It spends the first days finding out, showing your product to different shoppers, at different moments, and watching what happens. During this learning window the results are genuinely noisy: a slow first afternoon, a quiet weekday, or a run of curious clicks that do not buy can all drag the early numbers down for reasons that have nothing to do with whether the campaign is any good. A seller who reads those first two days as a verdict is reading random static as if it were the broadcast. This is why the strongest single rule of Meesho ad duration is also the simplest: give a campaign at least 7 days before you judge it. The week smooths the noise and lets the system reach the performance the catalog is actually capable of.
Killing a campaign in a day or two is not just premature; it is expensive. You have already paid to buy those early clicks and to start the learning process, and pulling the plug throws all of that away before it can return anything. Worse, you learn nothing you can use next time, because you never let a single campaign reach the stage where its numbers mean something. If you find yourself launching, killing and relaunching every few days, you are not testing; you are paying the learning cost over and over and never collecting the payoff.
What the week actually buys you
Seven days of data gives you three things a day or two cannot. First, it averages out the daily swings so you can see the campaign’s real level rather than a single lucky or unlucky afternoon. Second, it gives the ad system time to close its learning window and settle into steady delivery. Third, it produces enough orders, on most catalogs, to say something honest about cost-per-order and return after RTO and returns, which is the number that actually decides whether to keep going. Budget for the week up front, treating the early spend as the cost of information, and read the verdict once, at the end, not continuously along the way. For how to size that budget, see how much to spend on Meesho ads.
What a campaign looks like over its first week
The same campaign can look like a failure on day two and a winner on day seven. Here is the pattern, and how confidence in a verdict grows with the days you let it run.
When to let a campaign close itself, and when not to
The duration setting is a decision, not a default. Match it to what the campaign is for, a one-off test, a seasonal push, or an ongoing winner.
| Campaign purpose | Duration setting | Why |
|---|---|---|
| Testing a new catalog | Fixed window, at least 7 days | Long enough to clear the learning window, closed enough to force a decision |
| Proven, profitable winner | No end date | Keeps a validated earner running continuously so it never quietly stops |
| Festival or sale push | Fixed end date on the event | The campaign should close itself when the occasion ends |
| Clearing seasonal stock | Fixed end date at stock-out | No point advertising a catalog you will soon be out of |
| Scaling a week-old winner | No end date, raise budget in steps | Duration is settled; the lever now is budget, not the end date |
One decision a week, not one a day
After the first week, a steady weekly review keeps every campaign honest without letting daily noise pull you into constant, contradictory changes.
Three habits that keep durations sane
Commit to the week
Decide up front that you will not touch the campaign for 7 days except to fix something broken. Budget for the whole window so the early spend is planned, not a shock.
Set No end date on winners
Once a catalog has proven itself over a fair week, put it on No end date so it keeps earning continuously while you turn your attention to the next test.
Book a weekly review
Pick one day a week to read every campaign on seven days of data and make one decision each: scale, hold or pause. Skip the daily twitching.
Most bad duration decisions are impatience wearing the mask of decisiveness. Slowing down is the discipline that separates sellers who compound from sellers who churn.
Pausing and restarting is not a reset button
It is tempting to treat pause and restart as a clean slate: if a campaign is not working, stop it, wait, and start fresh. But restarting does not hand you a new, unspoilt campaign. It can wipe the learning the ad system built and drop the catalog back into the noisy early period, so you pay the learning cost a second time and are no wiser than before. If a campaign genuinely fails after a fair week, the right move is almost never to restart the same thing and hope; it is to change something real, the price, the first image, the listing quality, and then give the improved version another proper week. Pausing to fix is productive. Pausing to reset is a treadmill.
Winners deserve permanence, tests deserve deadlines
Once a catalog has earned its verdict over a week, the duration question flips. A proven, profitable winner should not be sitting on a short end date that quietly switches it off when you are not looking; it belongs on No end date so it keeps earning while you focus elsewhere, with the weekly review as your safety check. Tests and time-bound pushes are the opposite: a festival campaign or a stock-clearing push should carry a fixed end date so it closes itself on schedule, rather than lingering past the point where it makes sense. Getting these two settings right, permanence for winners, deadlines for experiments, is most of what good duration management actually is. The rest is patience, and a weekly review honest enough to act on. To go deeper on the levers around this, see auto CPC vs manual CPC and Meesho ads keyword targeting.
The mistakes to watch for
The single most common error. Early data is noise, not signal. A campaign that looks dead on day two frequently steadies by day seven, so a day-two verdict throws away both the learning window and the money already spent.
Daily numbers bounce for reasons that have nothing to do with the campaign's real quality. Chasing each up and down day leads to constant, contradictory changes that never let a campaign settle. Decide on the week, not the day.
A validated, profitable catalog left on a short end date stops earning the moment the date passes, often without you noticing. Proven winners belong on No end date so they keep running while you focus elsewhere.
Pausing and restarting does not give a fresh clean slate; it can wipe the learning a campaign built and drop it back into the noisy early period. Fix the listing or price instead of hitting restart and hoping.
Set-and-forget with no review is how losers quietly bleed budget for weeks. A campaign needs a week to prove itself, and then a weekly check to stay honest. Skip the review and you never learn which campaigns to scale or cut.
Sources & further reading
Duration settings and the learning behaviour of the ad system can change; always confirm the current options inside your own Meesho Supplier panel.
Robnu makes your weekly verdict trustworthy
Robnu does not start, stop or schedule your campaigns, and it never decides how long they run; that stays yours in the Meesho panel. What it does is make the profit you judge each campaign on real. It reconciles every Meesho settlement to the rupee, verifies the ad, RTO and return deductions, and flags the wrong ones, so a week of data adds up to a verdict you can act on rather than one the panel merely implied. It runs order processing end to end and scales from a single order a day to 50,000 and beyond.
Free for every seller right now, and forever free under 25 orders a day when paid pricing launches. See it on Meesho order management or the full order management system overview.
How long to run Meesho ads, answered
Give a Meesho campaign at least 7 days before you decide whether it works. The ad system needs a learning window to gather impressions, clicks and orders, and a single day or two rarely produces enough data to tell a real winner from a slow starter. Judging too early is the most common beginner mistake, because a campaign that looks weak on day two often steadies by day seven. Wait for the week, then read the numbers.
Because one or two days is almost never enough data to judge anything. Early results swing wildly: a quiet Monday, a slow first afternoon, or a handful of non-converting clicks can make a perfectly good campaign look like a failure. The ad algorithm is also still learning who to show your catalog to in those first days. Killing a campaign that early throws away the learning window and the money you already spent buying it, so you learn nothing and start over from zero.
The learning window is the early period, roughly the first several days, when the ad system is still working out which shoppers to show your catalog to and at what price. During this window results are noisy and often worse than the campaign's steady state, because the system has not yet settled. Giving it time to learn, rather than reacting to every daily swing, is what lets a campaign reach the performance it is actually capable of.
For a proven, profitable catalog, using No end date keeps a winner running continuously so it never stops earning while you are not watching. For a test, a promotion or a seasonal push, a fixed end date makes sense so the campaign closes itself on schedule. The rule of thumb is: end dates for experiments and time-bound pushes, No end date for the ongoing winners you have already validated over at least a week.
Review weekly. A week is long enough to smooth out daily noise and short enough to catch a campaign that has genuinely turned. A weekly rhythm lets you compare like with like, judge each campaign on seven days of data, and make one considered decision, scale, hold or pause, rather than twitching at every hourly fluctuation. Daily glances are fine for spotting something broken, but the real decisions belong to the weekly review.
Most campaigns need a few days just to exit the noisy early period, and about a week before their steady performance is readable. Some catalogs pick up almost immediately, others take the full learning window to find their audience. The safe assumption is that you will not know whether a campaign works until it has run at least 7 days, so plan and budget for that window rather than expecting a verdict overnight.
You can, but frequent pausing and restarting can unsettle the learning window and reset the momentum a campaign has built, so it is not free. If a campaign is genuinely losing money after a fair week of data, pausing it to fix the price, image or listing is sensible. But pausing out of impatience, before the learning window closes, tends to keep you stuck at the noisy early stage forever, because you never let a campaign mature.
No. Robnu does not start, stop, schedule or manage your ad campaigns; the duration and the end-date choice stay entirely with you in the Meesho panel. What Robnu does is make sure the profit you use to judge a campaign over its week is real, by reconciling every settlement to the rupee and flagging wrong ad, RTO and return deductions. A weekly review is only as trustworthy as the numbers behind it, and that is the part Robnu handles.
Related seller guides
More on the operations, money and claims that decide whether a marketplace catalogue actually makes money.
Meesho Smart Campaign Explained: How Smart Ads Work (2026)
Meesho Smart Campaign lets the platform auto-optimise catalog selection, budget and CPC for you. Learn how Smart Ads work, when to use them, and how they differ from manual campaigns.
How to Scale Meesho Ads: Increase Budget on Winners Safely
How to scale Meesho ads: once a campaign proves good ROI, raise the budget gradually, scale winners, pause losers, never double overnight, and watch ROI.
How to Optimize Meesho Ads: Track Orders, Cut Wasted Spend
The optimisation playbook: run seven days, track orders not clicks, pause losers, scale winners, and fix the pricing behind clicks with no sales.
How Many Images Per Meesho Listing? 4 to 5 Photos Explained (2026)
How many images per Meesho listing? Upload 4 to 5 high-quality photos from different angles and in use. One image is never enough. Here is the front, back, side and top set that sells.
Meesho Product Ads vs Brand Ads vs Smart Campaign (2026)
Meesho Product Ads promote single catalogs, Brand Ads promote your store, and Smart Campaigns auto-optimise. Compare all three ad types, how each is billed, and when to use which.
Meesho Ad Campaign Setup: Every Field Explained (2026)
Every field in the Meesho campaign builder decoded: name, catalogs, daily vs total budget, duration, recommended vs manual CPC, and the summary step.
How Much to Spend on Meesho Ads: A Beginner Budget Guide
How much to put behind Meesho ads: a starting daily budget, daily vs weekly vs custom, spending behind winners, and the budget that keeps ROI positive.
How to Get More Reviews on Meesho: The Compliant Way (2026)
How to get more reviews on Meesho without breaking policy: deliver a good product on time, add a polite non-incentivised feedback insert (can lift review rate about 15 percent), and never buy or in...