Mid-August Reset: What to Do With the Summer Inventory That Never Moved

Open the summer collection that never sold and pick the worst product in it. Now write down two numbers. How many people viewed that product page since June 1, and how many units you sold. If the first number is 41, you do not have a pricing problem. You have 41 people. No discount creates the 42nd.

Every August ends the same way. The summer assortment underperformed, fall inventory needs the shelf space and the cash, and the bulk markdown goes live because it looks decisive. Here is the part nobody says out loud. Most of that inventory did not fail on price. It failed on traffic. Which makes a mid-August markdown the most expensive possible way to discover that nobody ever saw the product.

By the end of this you can sort every unsold SKU into one of two piles using two numbers from your own product report. The Unseen and the Unwanted. Each pile gets its own plan, because they do not have the same problem. One of them needs a markdown. The other needs an audience. Start by splitting the pile.

Dead Inventory Is Two Problems Wearing One Costume

From the shelf, every unsold box looks the same. That is the trap. The Unwanted got in front of hundreds of shoppers and lost, so the price, the photos, or the product was wrong. The Unseen never got in front of anyone. It did not fail. It never got to compete. A bulk markdown is the right answer for exactly one of those piles. Most stores apply it to both.

The Diagnostic Is Two Numbers, Not a Meeting

You need two things per dead SKU. Views tell you whether the product got an audience. Add-to-cart rate tells you what that audience decided.

1. Pull views: product page views for each unsold SKU, June 1 to today.
2. Pull add-to-cart rate: carts divided by views over that same window.
3. Compare against yourself: not an industry benchmark, your own store average.

A product with 900 views and an add-to-cart rate a third of your store average was seen, considered, and passed on. That is real rejection. Rejection responds to price, photos, and copy. A product with 41 views over ten weeks was never in the running. Its conversion rate is a rounding error. So is any conclusion you draw about its price.

Why the Two Piles Look Identical From the Warehouse

Physical inventory has no memory of how it got there. The shelf cannot tell you whether 900 people said no or nobody ever showed up. So the decision defaults to price, the only lever a warehouse view offers. The difference between the piles lives one click deeper, in the product report you do not open every morning. That click is why the check gets skipped.

Dead inventory is not one pile. It is two. The Unwanted got in front of hundreds of shoppers and lost. The Unseen never got in front of anyone.

A Discount Can Move Your Conversion Rate. It Cannot Move Your View Count.

A discount acts on one number. It changes the share of people who see the product and decide to buy it. It does nothing at all to the number of people who see it. On a product with real traffic that is fine, because the audience is already there. On the Unseen pile, the audience is the entire problem, and price cannot touch it.

The Arithmetic on a Product Nobody Saw

Take a $60 item at a 55% gross margin. That is $33 of margin per unit and $27 of cost. Since June 1 it has 41 product page views and 2 units sold. You cut it 40%. The price is now $36, the cost is still $27, and margin per unit drops to $9. Now assume the markdown performs better than you had any right to expect and triples the conversion rate on that same traffic. Six units instead of two. Gross profit goes from $66 to $54. You tripled unit sales and made less money.

And you still have 60 units on the shelf. At six units per ten weeks, you sell the last one in the summer of 2028. That is three fall assortments from now, each one needing the space this pallet is standing in.

What This Changes About the Order of Operations

The mistake is not the discount. It is the sequence. Price is the last lever, not the first, because it is the only one that permanently costs you margin. On the Unseen pile every cheaper lever is still sitting untouched. The product is not in a collection anybody browses. It never showed up in internal search. It has no recommendation placement. It has never been in front of a paid or email audience. Testing price before testing distribution means paying for information you could have had free.

A discount raises the percentage of viewers who buy. It cannot raise the number of viewers. On the Unseen pile, that ceiling is the whole story.

Treat those cutoffs as starting points, not laws. Calibrate them against your own store average and your own season length.

August Hands You Traffic You Did Not Buy. Spend It on the Unseen.

Mid-August is the one stretch of the summer when sessions run above your June and July baseline. Back-to-school pulls people back into shopping mode, and some of them land on you. Now watch where that surplus goes. Almost all of it flows to the same six products that were already converting, because those are the ones sitting in your featured collection and your homepage row. The products that need traffic get none of the traffic that showed up free.

Distribution Is a Test, Not a Favor

Pick three Unseen SKUs and put them in the paths people are already walking. The collection getting the most sessions this week. The recommendation slots on your highest-traffic product pages. The spot a shopper passes on the way out. Give it two weeks at full price and watch add-to-cart rate, not revenue.

Either answer is worth having. If the product picks up a healthy add-to-cart rate once real traffic hits it, price was never the issue and you just saved yourself the markdown. If it gets 800 views and still nobody carts it, you have moved it into the Unwanted pile with evidence. The markdown in the next section becomes an informed decision instead of a guess.

Who Gets an Offer on a Rediscovered Product

More traffic brings the visitor question right back. A dedicated buyer who found the product, read the reviews, and is deciding between two colorways does not need a discount to finish. Hand them one and you paid for a sale you already had. The walk-away customer glancing at it on the way out is the person an offer is for. The point of putting a product in front of more people is not to discount it to more people.

This is where building the piles by hand starts to hurt. Growth Suite's product segmentation groups your catalog by how sessions actually move through each product, into buckets like Invisibles, Stoppers, Prospects, and Gems, so the Unseen pile stops being a spreadsheet you assemble on a Sunday. Its trending and frequently-bought-together widgets then give those products the placement they never had, themed to your store and rendered without flicker.

August gives you traffic you did not pay for. Most stores hand every session of it to the six products that were already fine.

The Unwanted Pile Deserves a Markdown, and a Fence Around It

For a genuinely rejected product, the discount is the right call. The question is not what percentage feels fair. It is the highest price that clears the units before the fall assortment needs the cash and the shelf.

Set the Exit Price Against Capital, Not Against Feelings

Work backwards from the shelf. Sixty units, and you need the space and the working capital by the last week of September. So the exit price is whatever moved sixty units in six weeks the last time you discounted something comparable. Pull that number from your own history, not from the percentage that sounds acceptable. Holding an Unwanted SKU at full price is not patience. You are paying storage and tied-up capital for the privilege of taking a deeper markdown in November.

Rotate a Few, Not the Whole Store

A permanent clearance collection does something quiet and expensive. Every full-price product standing next to it starts to look like it is next. Rotate a handful of markdowns at a time instead, with real end dates, and the store reads as a shelf that changes rather than a store that is always on sale. Then put a fence around it. Exclude new arrivals, specific vendors, and anything already discounted, or the deal you built for last summer's losers quietly lands on the fall pieces you have not sold a single unit of.

Growth Suite's product deals run exactly this shape. Native price changes with real timers, only a limited number of products live at once, and a cooldown before a SKU can come back around. Its discount rules build the fence, with exclusions by vendor, title, and new arrival status, plus a cap on how many dollars any single order can take off.

Two Piles, Two Plans, One Week

Unsold inventory is not one problem. It is the Unseen and the Unwanted, and a bulk markdown only fixes one of them. Two numbers from your product report sort the whole thing in about ten minutes. On a product with 41 views, tripling conversion still makes you less money than leaving the price alone, because a discount cannot raise the view count.

Here is the test for this week. Pick your three worst SKUs by capital tied up, sort your unsold products by page views, and find out how many of them were ever actually seen. The answer usually changes the plan. Then run the same check in week three next season instead of week ten, because week ten is the reason this pile forms every summer.

If you are about to mark down a summer collection without knowing which products anybody ever looked at, Growth Suite helps you tell walk-away customers apart from dedicated buyers, gives your overlooked products real placement, and rotates markdowns only on the ones that earned them. So you clear the shelf without discounting the shoppers who were already going to buy. It is free to install on the Shopify App Store, with a 14-day free trial.

Frequently Asked Questions