Solved: How Long Should a Facebook Ad Run Before Pausing?

Hi there,

Thanks for reaching out!

I've had a look over the situation you described with your new shop. It’s a common and frankly frustrating place to be in, so don't worry, you're not alone. Seeing people add your product to their cart but not follow through to buy is a classic problem. I'm happy to give you some initial thoughts and guidance based on my experience. There's quite a bit to unpack here, so I'll go into some detail for you.

We'll need to look at the checkout drop-off...

Okay, first things first. You said you're getting a 4-5% Add to Cart (ATC) rate. Tbh, that's not bad at all for a new store. It tells me a few things straight away: your product is likely desireable, your pricing on the product page isn't scaring people off, and your ads are reaching people who are at least somewhat interested. The video replays confirm this – 50% interaction is a good signal. So, the problem isn't getting people to the store or getting them to like the product. The big red flag is the drop-off between ATC and the final sale. This is where we need to put on our detective hat.

When we see a funnel break down at this specific point, it's almost always down to a handful of culprits. You need to go through your checkout process yourself, with a brutally honest eye, and see what might be tripping people up.

-> Unexpected Costs: This is the number one killer of conversions. A customer has mentally agreed to pay, say, £50 for your custom product. They go to checkout, and suddenly you're asking for £10 for shippin'. That's a 20% price hike they weren't expecting. It feels like a bait-and-switch, even if it's not intentional. It breaks trust and they leave. You should be upfront about shipping costs, either by offering free shipping (and building the cost into your product price) or by having a clear banner on the site that states your shipping rates.

-> A Comlicated Checkout Process: How many fields are you asking them to fill in? Do they have to create an account to buy? Every extra click or piece of information you ask for is another chance for them to get distracted or give up. You want this process to be as frictionless as possible. Use things like Shopify Pay, Apple Pay, Google Pay, PayPal – one-click options are your best friend here. If someone can buy without having to find their credit card and type in 16 digits and their full address, your conversion rate will thank you for it.

-> Lack of Trust: You're a new shop. People are handing over their hard-earned money to a complete stranger online. Your website needs to scream "I am legitimate and you can trust me." This is often overlooked but it's massive. Do you have clear contact information? A physical address if applicable? A well-written 'About Us' page that tells your story? Customer reviews or testimonials (even if you have to ask early customers for them) are huge. Trust badges for secure payments can also help. I remember looking at one client's store, an ecommerce business selling handcrafted goods, and their page was just a bit cluttered and didn't feel solid. Adding some testimonials and making the contact info more prominent made a noticable difference before we even touched the ads.

Looking at your store's analytics is good, but you also need to think about the psychology of the buyer at that final step. They're on the verge of committing, and any little bit of doubt or friction can push them away. Fix this part of your funnel first, because sending more traffic to a leaky bucket is just a waste of money.

I'd say you need a solid retargeting strategy...

Right, so let's talk about those people who added to cart and left. These people are not a lost cause. They are your most valuable audience. They've raised their hand and said, "I want this product." Something stopped them, sure, but the interest is there. Leaving them be is like a shopkeeper watching a customer put an item on the counter and walk out, and not even bothering to ask if they need any help. It's madness.

You need to be running dedicated retargeting campaigns. This isn't optional for ecommerce; it's fundamental. This is where a big chunk of your sales will come from, especially for a considered purchase like a custom product.

This is where understanding the marketing funnel comes in handy. I usually structure my accounts to reflect this. You have your "cold" traffic at the Top of the Funnel (ToFu), and then your "warm" and "hot" audiences in the Middle (MoFu) and Bottom (BoFu).

Your cart abandoners are pure BoFu. They are as hot a lead as you can get without a purchase. You should have a seperate ad set, or even a whole campaign, dedicated just to them. Here’s a basic structure I'd prioritise:

You can see how this works. You're not showing the same ad to everyone. You're continuing the conversation based on how interested they've shown they are. For your BoFu audience, you can be direct. For MoFu, you need to nurture them a bit more. One campaign we worked on for a women's apparel brand saw a huge part of their 691% return on ad spend come from a really aggressive and well-structured retargeting funnel on Meta Ads. People rarely buy clothes on the first visit; they need reminders. Your custom product is likely the same.

You probably should stop tweaking your ads so often...

This is a big one, and probably the most direct answer to your question. You said you keep tweaking the ads every 2-3 days. You have to stop doing this. I know it's tempting. You're not seeing sales, so your instinct is to change something, anything. But you're actually shooting yourself in the foot.

Meta's advertising algorithm is a powerful machine, but it needs data and time to work. Every time you create a new ad or significantly edit an ad set (changing the budget, targeting, or creative), it enters what's called the "Learning Phase". During this phase, the system is actively exploring the best way to deliver your ads – figuring out who is most likely to convert, what time of day is best, etc. It needs to get around 50 conversions per ad set per week to exit this phase and become stable. If you're only getting one sale, you're nowhere near that.

By tweaking things every couple of days, you are constantly resetting the learning phase. Your ads never get a chance to optimise properly. You're keeping them in a permanent state of inefficiency. It's like trying to grow a plant but pulling it out of the soil every day to check the roots. You have to let it be.

So, how long should you let an ad run? There's no magic number of days. The rule of thumb I use is based on spend. You need to decide what you're willing to pay for a sale – your target Cost Per Acquisition (CPA). Let's say your product costs £50 and you're happy to spend £20 to get one sale. A good starting point is to let an ad set spend at least 1.5x to 2x your target CPA before you even think about turning it off. So in this case, you'd let it spend £30-£40. If it has spent that much and has zero sales, then okay, it's probably a dud. But turning it off after 2 days and £5 of spend tells you absolutely nothing. That one sale could have been just around the corner.

You need to be more patient and systematic. Set up your campaigns, define your audiences and creatives, and then let them run. Analyse the data on a weekly basis, not daily. This discipline is what seperates amateurs from professionals in this game.

You'll need to expand your targeting thoughtfully...

You mentioned you've been using targeted ads but haven't tried broad. This is another area with huge potential for you. A common mistake I see is people either going way too specific with their targeting, reaching a tiny audience, or way too broad from the start with no data. You need a structured approach to testing.

I always prioritise audiences based on their likely performance. It's a hierarchy of quality.

META ADS AUDIENCE PRIORITISATION (for a new store)

-> 1. Retargeting (BoFu/MoFu): We've already covered this. This is your number one priority because it's your warmest audience. People who've visited your site, viewed products, and added to cart.

-> 2. Lookalike Audiences: Once you have enough data (you need at least 100 people in a source audience), you can create Lookalikes. This is where you ask Meta to find new people who are similar to your existing customers or visitors. A Lookalike of your purchasers is gold dust. A Lookalike of your Add to Cart audience is the next best thing. Then product viewers, then all website visitors. You'd test these in order of value. I've seen Lookalikes of high-value customers completely transform campaigns for SaaS clients, and the principle is identical for ecommerce.

-> 3. Detailed Targeting (Interests/Behaviours): This is your primary tool for cold (ToFu) traffic right now. The key here, as you've found, is to be specific, but not *too* specific. Think about the profile of your ideal customer. What magazines do they read? What brands do they love? What hobbies do they have that relate to your "custom made product"? If you sell custom leather wallets, you might target interests like "craftsmanship," "leatherworking," brands like Filson or Saddleback Leather, and magazines like The Rake. The mistake is targeting something too broad like "fashion" or "shopping." You need to find interests that your ideal customer is much more likely to have than the general population. Make a few different ad sets, each with a tight 'theme' of related interests, and test them against each other.

-> 4. Broad Targeting: This is where you literally just set the age, gender, and location, and let Meta's algorithm do the rest. You should *not* do this yet. Broad targeting can work incredibly well, but only when your pixel has thousands of conversion events to learn from. It needs a huge amount of data to understand who your customer is. For a new store, it's like flying blind. Stick to detailed targeting and lookalikes for your cold traffic for now.

I'd suggest setting up a simple ToFu campaign with 3-4 ad sets. Each ad set would target a different interest theme. Let them run (with a modest daily budget each) for a week and see which one performs best in terms of ATC and, hopefully, sales. Then you can start to scale the winner.

We'll need to look at your creative approach...

Finally, let's talk about the ads themselves. You have a "custom made product". This is a huge advantage you need to press. You're not selling a generic, dropshipped widget. You're selling something unique, with a story. Your ads must reflect this.

Are you just using static images of the product on a white background? That's fine for a retargeting ad maybe, but for cold traffic, you need to grab attention and build value.

-> Video is King: For a custom product, video is a must. Show it being made. Show the detail, the craftsmanship, the materials. This builds immense perceived value. A 30-second video showing the hands that make the product can be more powerful than any sales copy.

-> Show, Don't Just Tell: Show the product in use. In a lifestyle context. Help the customer imagine it in their own life. If it's a piece of custom jewellery, show it on a model at a nice event. If it's a custom piece of home decor, show it in a beautifully styled room. You're not selling a product; you're selling the outcome, the feeling, the status that product gives someone.

-> User-Generated Content (UGC): Once you have some sales, get your customers to send you photos or videos of them with your product. UGC is incredibly powerful. It's authentic social proof. An ad featuring a real, happy customer is often far more effective than a slick, professional ad. For a software client, we found that simple UGC-style videos dramatically outperformed polished corporate videos. People trust people.

You should be split-testing your creative constantly. In your ToFu campaign, test a video ad vs. a carousel ad vs. a strong static image. See what resonates with a cold audience. Then, in your retargeting campaigns, you can use different creative that speaks more directly to them.

This is the main advice I have for you:

I know this is a lot to take in. The key takeaway is to be more systematic and patient. You have good initial signals (the ATC rate), so you're not starting from zero. You just need to fix the leaks in your funnel and refine your advertising methodolgy.

Running a successful paid advertising operation is a lot more involved than just boosting a post. It's about building a robust, full-funnel system that acquires customers efficiently and predictably. It involves understanding data, user psychology, and the technical nuances of the ad platforms. It takes time to build this system and expertise to manage it effectively.

We've worked with numerous ecommerce clients, from subscription boxes where we achieved a 1000% return on Meta Ads, to cleaning products that saw a 633% return on Meta Ads, and we do it by implementing these kinds of rigorous, systematic processes. It's not about finding one 'magic' ad; it's about building a machine that works.

If you'd like to chat through this in more detail and have us take a direct look at your setup, we offer a free initial consultation. It might be helpful to have an experienced pair of eyes on it to spot the specific opportunities you could be missing.

Hope this helps give you a clearer path forward!

Regards,

Team @ Lukas Holschuh