Retailers do not discount randomly. Most run repeating markdown cycles driven by inventory turns, seasonal resets, and fiscal calendars, which means the same category tends to reach its deepest discount at roughly the same point each cycle. The practical consequence for arbitrage sellers is that the most common invisible mistake is buying a good product at the wrong point in the cycle: paying 30 percent off in week one when the same item reaches 60 percent off in week four. Sourcing on a schedule rather than on impulse fixes more margin than finding better products does. This piece covers how cycles work, how to read one, and when waiting is wrong.
Disclosure: OAList Pro sells predicted sale windows, so I have an obvious interest here. The method below works whether or not you use a tool for it.
Why timing beats product selection
Almost all sourcing advice is about which product to buy. Very little is about when. That is backwards relative to how much money each decision moves.
Consider the same item at three points in a cycle. At first markdown, maybe 25 percent off, your ROI is thin and you probably pass. At second markdown, 40 percent off, the deal works and you buy. At final clearance, 65 percent off, the deal is excellent but only broken sizes remain.
Same product, same Amazon listing, same fees. Three completely different outcomes, decided entirely by when you showed up.
Now multiply that across a year of buying. A seller who consistently buys one markdown stage later than another seller is not slightly behind, they are running a structurally different margin profile on identical products.
How markdown cycles actually work
The mechanics are boring, which is why they are reliable.
Inventory turns. Retailers need shelf space for incoming stock. Product that has not sold by a certain age gets marked down on a schedule, and that schedule is usually policy rather than judgment.
Seasonal resets. Categories have hard end dates. Summer goods have to be gone before autumn resets. The markdown accelerates as the deadline approaches, which is why end-of-season clearance is deep and predictable.
Fiscal calendars. Quarter ends and inventory counts create pressure to move stock. Retailers with a January fiscal year end behave differently in December than those with a June year end.
Promotional calendars. Recurring events repeat annually within a fairly tight window, and the categories featured tend to repeat too.
The useful insight is that these are operational constraints, not marketing whims. A retailer marks down because holding the inventory costs money. That is why the pattern repeats.
Reading a retailer's cycle yourself
You do not need software to start. You need a record.
- Pick five retailers you actually buy from. Not fifty. Five you know.
- Log every deep discount you see. Date, category, depth. A spreadsheet is fine.
- After two or three months, look for interval and depth. How many weeks between deep markdown events. How deep the deep ones go. Which categories lead and which lag.
- Write down the predicted next window. Then check whether you were right, and adjust.
Within a quarter you will have real intelligence on five retailers, which is more than most sellers have on any.
The limit is obvious: this does not scale past a handful of retailers, and the categories drift. Doing it across 100 retailers by hand is not a project, it is a job.
What that looks like at scale
This is the gap OAList Pro was built for. The sale-cycle layer is modelled on roughly 27,000 recorded deal observations across 848 retailers, and it outputs a predicted next-sale window per retailer along with how deep that retailer typically goes and how often.
The behaviour change is the point. Instead of opening your inbox and reacting to whatever arrived, you look at which retailers are entering their discount window this week and prepare capital and gating in advance.
Elite adds alerting on those predicted windows, so a retailer entering its window pings you rather than requiring you to check. Pricing across the tiers is $29, $89, and $239 per month, or $228, $708, and $1,908 annually, which is $19, $59, and $159 per month on annual billing. Basic and Pro include a 7 day free trial, Elite has a 30 day money-back guarantee.
The honest limitation: a predicted window is a probability, not a promise. Retailers change strategy, supply chains break, and a cycle that held for two years can shift. Treat it as a strong prior that tells you where to point attention, not a guarantee.
When waiting is the wrong call
Timing discipline turns into paralysis if you apply it blindly.
Limited quantity. If there are 40 units nationwide, the next markdown is theoretical. Buy now or lose it.
Products with many competing buyers. If a deal is already circulating, waiting means arriving after the stock is gone.
Seasonal windows you cannot miss. A Q4 product bought in December at a better discount may still be worse, because you missed the selling window. Depth of discount is worthless if the units land after demand ends.
When your capital is idle anyway. A 30 percent ROI deal today beats a hypothetical 50 percent deal in three weeks if the alternative is cash sitting still. Money doing nothing has a cost people forget to count.
The rule I use: wait when supply is deep and the deadline is soft. Buy now when supply is thin or the deadline is hard.
Building the schedule into your week
Pick sourcing days rather than sourcing constantly. Cycle awareness only pays if you act on it, and acting requires having capital ready when a window opens rather than spent on whatever appeared on Tuesday.
Hold reserve capital for windows. Sellers who are fully deployed cannot act when the best window arrives. Keeping a portion unspent is not caution, it is positioning.
Sort gating before the window, not during. Getting ungated in a category takes time you will not have when the sale starts. Do it in the quiet weeks.
Log what actually happened. The prediction is only as good as your feedback loop. Note when a window hit, when it missed, and how deep it went.
Where to go next
For which retailers reward this approach most, see the best retailers for online arbitrage. For the seasonal version of the same logic, Q4 sourcing covers the hard deadlines. And if you are weighing whether a deal feed belongs in your stack at all, the lead list question is the more fundamental one to answer first.
Know where to source, and when
OAList sends classified arbitrage deals across 100+ retailers every morning, and Pro predicts when each retailer runs its next deep sale. Start with a 7 day free trial.
Start my free trialFrequently asked
What is a retailer sale cycle?
A retailer sale cycle is the repeating pattern in how often and how deeply a specific retailer discounts. Most retailers do not discount randomly. They run predictable markdown schedules tied to inventory turns, seasonal resets, and fiscal calendars, so the same category tends to hit its deepest discount at roughly the same point each cycle.
How do I know when a retailer will run its next deep sale?
Track the retailer's discount history over time and look for the interval between deep markdown events and the typical depth of each. Doing this by hand across many retailers is impractical, which is why OAList Pro models it from about 27,000 recorded deal observations across 848 retailers and predicts a likely next-sale window per retailer.
Is it better to buy at first markdown or wait for clearance?
It depends on whether stock will survive. First markdown gives you certainty of supply at thinner margin. Final clearance gives you the best margin on whatever is left, which is often the wrong sizes, colours, or variations. For products with deep stock, waiting usually wins. For limited-quantity items, first markdown is often the only real chance.
Do sale cycles still work if everyone knows about them?
More than lead lists do, because timing intelligence tells you when to look rather than handing everyone the same product. Two sellers who know a retailer discounts deeply in a given window will still buy different products based on their own categories, capital, and gating. Shared timing does not collapse margin the way a shared product list does.