Product launch marketing: the complete guide

Write down what a successful launch looks like in numbers before you do anything else. Without that, every result is arguable, and a launch that quietly failed gets described as a learning experience instead of being fixed.

Launches usually have one of three jobs, and they need different plans. The first job is proving demand: you want to know whether people will pay for this at all, so the goal is a certain number of orders from people who did not already love you, and the plan should include cold traffic. The second job is raising the average order: the new product exists to attach to something you already sell, so success is measured by attach rate and by whether orders that include it are worth more. The third job is bringing customers back: the product is a reason to email people who bought once, and success is a repeat purchase rate rather than new customer count.

Pick one. A product that has to do all three usually gets a launch that does none of them well, because the audiences and the messages pull in different directions.

Then set the numbers. How many units do you need to sell for the launch to be worth the stock you bought. What is the most you can pay to acquire a first buyer, given the contribution margin on this specific product and what you know about how often people buy again. How long is the window you are judging: a week is usually too short for anything but a hyped drop, and a quarter is usually too long to act on.

Write those figures on one page and share them with everyone involved. When launch week gets loud and someone wants to add a bigger discount or a new channel, the page is what you check the idea against.

Before you promote a new product, check what it does to the ones you already sell. New products often move revenue around the catalogue instead of adding to it, and a launch that looks strong on its own can be flat or negative once you count what stopped selling.

Start by naming the product it is closest to. If the new item sits in the same category, at a similar price, solving the same problem, expect some of its sales to come out of the existing product rather than from new demand. That is not always bad. A replacement with better margin or a better repeat rate is worth the swap. A replacement with the same margin that splits your ad budget across two listings usually is not.

Then look at the practical effects. Two similar products in the same shopping campaign compete for the same searches, and the platform will pick whichever has more data, which is usually the old one. Two similar products on a collection page split attention and make the choice harder, which lowers conversion for both. Two similar products in your feed can trigger duplicate content and near identical titles.

Decide in advance how you will tell the difference between addition and movement. The straightforward version is to watch the category total rather than the product line. If the new product sells and the category grows, you added demand. If the new product sells and the category is flat, you moved it.

Finally, plan the merchandising. Where does this product sit in the navigation, on which collection pages, and in what position. What does it get compared against on the page. If it replaces something, decide what happens to the old page: kept for the existing customers who reorder, redirected once stock runs out, or left live with a clear pointer to the new version. Making that decision now saves a messy cleanup later.

The single biggest difference between launches that work and launches that do not is whether there was an audience waiting. Build it before the product is ready, because a launch to nobody is just a new page on your site.

Start with the people you already have. Segment your list by what they bought and what they browsed, and identify who this product is genuinely for. A launch email to your whole list gets a middling response and costs you unsubscribes. The same email to the segment that bought the related product or looked at that category performs better and annoys fewer people.

Then build the waiting list. Put a signup on a simple page describing what is coming, and use every place a curious person already lands: your social profiles, the footer of your order confirmation, a mention in your regular sends, and any organic traffic you get for related terms. Ask for as little as possible. An email address, and a phone number only if you can use it properly.

Give the waiting list something for waiting. Early access works better than a discount because it does not set a price expectation, and because the people who want to be first are the ones who will buy. If you have a limited first run, saying so honestly is the strongest reason to sign up, provided it is true.

Use the wait to learn. Ask the people who sign up what they want to know, in one question. The answers become the product page copy, the ad angles and the FAQ. This is free research from the exact people you are about to sell to, and it usually reveals a question you had not thought to answer.

If you run paid ads, spend a small amount in this window building audiences rather than chasing signups at any cost. Video views, page visits and engagement collected now give the launch campaigns something to work with on day one.

Every visitor a launch produces ends up on one page, so write that page before you write a single ad. A launch with excellent ads and a thin page spends money teaching people to leave.

A new product has a harder job than an established one. It has no reviews, no social proof, and nobody has heard of it. The page has to carry all of that on its own. Start with the questions a first buyer asks, in the order they ask them: what is it, who is it for, what does it do that the alternative does not, what does it cost to get it to me, how long will it take, what if it is wrong.

Put the answers where they are seen on a phone. The first screen needs the name, a photo that shows the product clearly, the price, the delivery expectation, and one line that says what it is for. Everything else can be below, but nothing important should be hidden behind a tab that most people never open.

Photography matters more at launch than at any other time, because there are no customer photos yet. Show scale next to something familiar, show it in use, show what is in the box, and show the details people will worry about. Video and a short demonstration do a lot of work when nobody has seen the product before.

Handle the proof gap directly. If you have no reviews, use what you do have: the reason you made it, how it was tested, the return policy that takes the risk off the buyer, and any early feedback you collected honestly from the people who tried it. Make the return terms easy to find. A clear return policy is the substitute for reviews on day one.

Finally, get the page live and indexed before launch day if you can, even if the buy button is replaced by a notify me option. Search engines and shoppers both benefit from the page existing before the traffic arrives.

Pick a single launch offer, work out what it takes from your margin, and stop there. Stacking a discount, free shipping and a gift on top of each other gives away more than the launch is worth and makes the normal price feel wrong afterwards.

Decide the everyday price first, then decide whether the launch price is different. A launch discount is not automatically the right move. It anchors the product low, and if the first buyers are your best customers, you have discounted the people who would have paid full price. For many stores the better launch offer is something that does not touch the price: free shipping on the new product, a bundle with an existing item, a bonus that costs you less than a discount would, or simply being first while stock is limited.

If you do discount, put a real end on it and honour that end. A launch price that never expires is just your price, and shoppers work that out quickly.

Run the margin arithmetic before you commit. Say the new product sells at 45 and costs you 22 once goods, fees, packing and the shipping you absorb are counted. Contribution is 23. If you offer 10 off, contribution drops to 13, which means you can pay far less to acquire each buyer and still be ahead. That change in what you can afford to spend should be decided before you set ad budgets, not discovered afterwards.

Think about the second order too. If this product is consumable or fits a subscription, the first order can afford to be thinner because the customer relationship is what you are buying. If it is a one time purchase with nothing obvious to follow, the first order has to pay for itself.

Write the offer, the end date, the exclusions and the margin figures in the same place as your launch targets, so the decision is documented rather than remembered.

Open the launch to the people who already know you, in a planned order, before you spend anything on reaching new people. Owned sends cost almost nothing, they give you the first orders and the first reviews, and they tell you whether the offer works before you put money behind it.

A workable sequence has four moments. First, a heads up to the segment most likely to want it, sent a few days ahead, saying what is coming and when. Second, the early access send to the waiting list and best customers, opening the product to them first. Third, the general announcement to the wider relevant segments. Fourth, a follow up to people who opened or clicked but did not buy, answering the objection rather than repeating the announcement.

Space those out. Sending all four in two days makes the launch feel like pressure. Spreading them over a week or two gives you time to read what is happening and change the message, which is the main advantage of selling to your own list first.

Write each send for its audience. Past buyers of the related product need a different opening than someone who joined the list last month. People on the waiting list already know what it is, so their email should be about getting it rather than explaining it. Reuse of one message across all segments is the fastest way to make a launch feel generic.

Update the automated flows at the same time. The welcome sequence should mention the new product where it fits. The post purchase flow for the related product is a natural place to introduce it. Browse and cart abandonment for the new page will need copy that works without reviews.

Use the first responses as material. The questions people reply with, and the reasons the non buyers give, are the exact objections your ads and page need to handle when you move to paid traffic.

Turn paid on after your own audience has bought, and start with the smallest, warmest audiences before spending on people who have never heard of you. A new product has no conversion history, so the platforms have nothing to work with, and broad spending in the first days mostly buys expensive learning.

Begin with remarketing and customer lists. People who visited the new page, people on the waiting list who did not order, and past buyers of the related product. These convert at a rate that gives the campaign real data quickly, and that data is what makes wider targeting work later.

Then widen in steps. Add lookalike or similar audiences built from the buyers you now have. Add interest or search targeting around the problem the product solves rather than the product name, because nobody is searching for a product name they have never seen. Keep each step separate for long enough to judge it.

On the search side, treat the new product as a small campaign of its own at first. Putting a brand new item into an existing campaign with established products usually means the platform spends on what already works and the new one never gets impressions. Give it its own budget so it has to prove itself, and use a product label in the feed so you can group and control it.

Run more than one creative angle from the start. You do not know yet which reason to buy will land, and the launch is your chance to find out cheaply. Three or four distinct angles, not three versions of the same angle with different colours.

Set a review point. Decide before you start what you will look at after the first week: cost per order, add to cart rate on the new page, and which angle produced the cheapest orders. Then act on it rather than letting a launch campaign run out of curiosity.

Decide in advance what you will do if nothing sells, and what you will do if everything sells in a day. Both happen, both are recoverable, and both go badly when the decision is made in the moment.

If it sells out, the risk is turning a good day into a bad month. Have a back in stock signup ready on the page before launch, so demand you cannot fill turns into a list rather than a bounce. Know your restock lead time and publish it rather than leaving the page saying sold out with no information. Pause the ads pointing at it, because paid traffic to an unavailable product is money spent on disappointment. Decide whether you will take pre-orders, and if you do, be specific about the ship date and honour it.

If nothing sells, resist the urge to discount on day three. Silence has causes, and the discount hides them. Work through the sequence: did people see it, did they click, did they land on the page, did they add to cart, did they check out. Each stage points at a different problem. No impressions is a targeting or budget problem. Clicks but no add to cart is a page problem: photos, price, or an unanswered question. Add to cart but no purchase is usually shipping cost, delivery time or trust.

Ask people directly. A short message to the waiting list members who did not buy, asking what stopped them, gets more useful answers than any amount of analytics. Most stores get the same two or three reasons back, and they are usually fixable on the page.

Have a stop rule as well. If the product has had a fair test against the targets you set and has not worked, decide what happens: reposition it, bundle it with something that does sell, or clear the stock and move on. A product kept on life support absorbs attention that the rest of the catalogue needs.

Launch day revenue tells you how good your email list is, not how good the product is. Judge the launch on the four numbers that predict whether it will keep selling once the announcement is over.

The first is the conversion rate of the new product page compared with your established pages. If the new page converts far worse on similar traffic, the problem is the page or the offer, and no amount of extra traffic fixes it. If it converts as well or better, you have something worth spending on.

The second is cost per order from cold traffic. Sales to your own list say nothing about whether this product can be advertised profitably. What matters is what a first order costs when it comes from someone who did not know you, measured against the contribution margin you calculated earlier.

The third is the effect on the rest of the catalogue. Look at the category total and at the product this one sits closest to. Growth in the category means new demand. A flat category with a strong new product means you moved sales rather than added them.

The fourth is what happens after the first order. Return rate tells you whether the page described the product honestly. Review sentiment tells you which parts to lead with next time. If it is consumable, the reorder rate over the following weeks tells you what the product is really worth, and whether you can afford to spend more to acquire the next buyer.

Write these four down at the end of the first month with a short note on each, then decide one thing to change. Launches produce a lot of information and most of it goes unused because nobody writes it down while it is still fresh.

Most products get one launch and then get forgotten, which wastes the assets the launch created. Plan a second wave while the first one is still running, so the product keeps selling after the announcement stops.

The second wave has something the first did not: proof. You now have buyers, photos from customers, reviews, questions people actually asked, and data on which angle worked. All of that changes the page and the ads. Update the product page with the review content and the answers to the real questions. Replace launch creative with the angle that produced the cheapest orders and with customer content where you have it.

Build the product into the places that sell without being asked. Add it to the relevant collection pages and to your navigation properly rather than leaving it under a new arrivals label forever. Add it to the post purchase flow of the product it pairs with. Add it to bundles. Put it in the feed with the right labels so it is included in shopping campaigns rather than sitting outside them.

Write the search side. Once you know the words customers use for it, which usually differ from what you called it internally, use those words in the page title, the description and the feed title. A launch names a product; the second wave names it the way shoppers search for it.

Then set a date to look again. A quarter after launch, check whether the product is still selling without promotion, whether it is being reordered, and whether it deserves more inventory or less. That review is what turns a launch into a line that earns its place in the catalogue, rather than a busy week you remember fondly.