The most honest thing an agency can tell you: you are not ready.

GGood ads amplify a product that works. They do not fix one that does not.

That is the rule. You learn it from this article, or from EUR 50,000 thrown out the window. The choice is yours.

If you have been through two or three agencies and none pulled a profit, stop. Ask yourself one question: were all of them bad? Or is the problem deeper than the ad account?

Maybe the problem is the product. The offer. The price. The site. Something no agency has the mandate to fix and no ad budget solves.

The line between the agency's fault and yours as a founder gets drawn in 20 minutes, if you look at the right numbers. It is not about feeling stupid. It is about knowing what to work on. Because if you throw EUR 5,000 a month at ads for a product that does not sell, you are not doing marketing. You are making donations.

There are founders who spend, collectively, over EUR 100,000 on ads before realizing the product was the problem. With that money they could have remade the product three times, tested ten offer variations, done real paid-customer research. Instead, they paid agencies to send traffic to a page nobody wanted to buy from.

Four signals the problem is the product, not the ads

1. Good CTR, bad conversion rate

CTR above 2% on Meta means the ad pulls. People click, they are curious, they liked something. But after the click, on-site conversion rate below 1%? The ad did its job. What happens after the click is not the agency's responsibility. It is yours.

Picture an account with 3.2% CTR and 0.3% conversion rate. The ad is excellent. Clean visuals, direct copy, relevant audience. The landing is a disaster. 7-second load time on mobile. Blurry photos pulled from the supplier's site. Zero reviews. Zero visible guarantees. The agency brings people to the door. The site chases them away in the first 3 seconds.

Here is how you check: take the average CTR for the last 30 days from Meta Ads Manager. If it is above 1.5%, the ad is working. Then take conversion rate from Shopify or GA4 over the same period. Below 1%, the problem is not the campaign. It is on your site.

2. Good add-to-cart, low purchase

Add-to-cart above 5% and Purchase below 1% of sessions. People want the product. They picked it. They put it in the cart. Then they stopped.

Why? Shipping fee that appears suddenly at checkout. EUR 35 on top of an EUR 80 product. Or card is the only payment option, and half of Romania's online shoppers still want cash on delivery. Or the checkout has 6 useless fields and on mobile you scroll three times.

Picture an account where 12% of sessions add to cart and 0.8% complete a purchase. You check the checkout. Shipping shows up only at step 4, costs RON35, and the only payment method is online card. And in Romania a good share of online buyers still pay cash on delivery. Customer adds to cart, sees they cannot pay the way they want, leaves. That is not the ad's fault. That is the checkout's fault.

Easy fix: show shipping cost on the product page. Add cash on delivery. Cut checkout to 3 steps max. These tests take two days, and the impact shows in the first week. A store that adds cash on delivery can lift conversion rate from under 1% toward 2% in a few days. Without changing a single ad.

3. Returns above 10%

If more than 10% of orders come back, the product does not deliver what the ad promises. One of three things: you oversell in copy, the site description does not match what comes out of the box, or quality is genuinely below expectation. Whichever it is, no ads team fixes that.

Picture a store with 18% returns. Products look great in photos, professional, well-lit. In reality, the material feels cheap to the touch. Customers order based on the photos, unbox, return. The cost of a return is not just the courier both ways. It is the product that most of the time cannot be resold at full price. It is margin disappearing. And if you do not measure it, you think you have 40% margin when you actually have 28%.

The number to check: return rate over the last 90 days, per product, not per store. If one product has 15% returns and another has 3%, stop promoting the first. Pull it from campaigns, fix the quality or align the copy expectations with reality, then put it back.

4. Nobody comes back

For repeat-purchase products (cosmetics, food, supplements, consumables), a repeat rate below 5% at 6 months is a clear signal. You sell once, then silence. Ads bring people on the first order. They cannot force them to love the product the second time. That is on you, on the product, on the post-delivery experience.

Think of a cosmetics brand with RON45 CAC and RON52 LTV. Looks fine at first glance. But if LTV comes after 9 months, and most customers never order a second time, then in reality most are only worth the first RON60 order at 30% margin. That is RON18 profit. Minus RON45 CAC. A loss of RON27 per customer. Ads bring buyers. The product does not convince them to come back.

Check it in Shopify: Reports, Returning customer rate over the last 6 months. If you sell consumables and under 10% come back, you do not have an ads problem. You have a product or post-sale experience problem. Maybe the packaging is weak. Maybe the formula changed. Maybe nobody writes to them after the first order to say thanks. Each of those is fixable. But you have to know it exists.

Five signals the problem is the agency

Flip side. If you see the numbers below, stop blaming the product. The agency is not doing its job. And here you have to be just as honest with yourself as when checking the product. Not all failures are your fault. Sometimes the agency really is weak.

• CTR under 1% on ad sets for over 2 weeks. The creative is weak or the audience is wrong. Or both. A CTR under 1% on Meta means the ad attracts nobody. People see it and scroll right past.
• Frequency above 5 within 7 days. They are not refreshing creative. People are tired of the same images. If someone sees the same ad 5 times in one week, they do not click. They start disliking it.
• CPA keeps rising month over month with no concrete explanation in the report. They are not testing, not optimizing. They raise bids to compensate for declining performance. It is the death spiral: spend more, results drop, spend even more.
• Same campaigns running for 3 months without a single new test. Zero strategy. Zero learning. Just the play button left pressed.
• POAS below 0.8 sustained for 4 weeks. They are scaling losses, not profit. ROAS 2.0 sounds good in the report. POAS 0.7 means you lose money on every sale. If the agency does not show you POAS, ask. If they do not know what it is, switch.

If you see these signals, the problem is the agency. Not the product, not the market, not seasonality. Someone is not doing their job. And switching is justified. But make sure the product is fine first. There are founders who switch four agencies when the problem was actually the checkout. Each new agency tries the same thing and fails the same way.

Third bucket: product is fine, offer is not

This is the zone most founders miss. The product works. The agency does its job. But the offer on the page is built to lose money. It is a presentation problem, not a substance problem. And that is exactly why it is frustrating: it is the easiest one to fix.

• Price above market with no justification visible in the first 3 seconds on the page. If you are 20% more expensive than the competition, the buyer needs to see why the moment they land. Superior quality? Longer warranty? Faster delivery? Something concrete. Otherwise, you are just more expensive.
• No guarantees, no extended returns, in a market where every competitor offers 14 or 30 days. Customers buy more easily when they know they can return without hassle.
• No bundles, no cross-sell. You leave AOV on the table on every order that could have been 30-40% bigger. Sell a face cream at EUR 80 and offer no 3-product set at EUR 180, and you are leaving money on every single sale.
• Shipping that pops up at checkout. EUR 35 on an EUR 80 item. Abandonment spikes instantly. A large share of checkout abandonments come from unexpected costs. Show them on the product page.
• Card only for payment, in a market where half of e-commerce in Romania uses cash on delivery. It is not about whether you like cash on delivery. It is that half your customers leave if you do not offer it. Math beats personal preference.

These five points get fixed in 1-2 weeks. You do not need to change the product. You need to change how you present it and how you sell it. Many founders skip this because they are too close to the product. They see what they want to see. Not what the customer sees when they arrive on the page for the first time and have 8 seconds to decide whether to stay.

What an agency that respects you looks like

It tells you in month one when the fault is not theirs. No spin, no dancing around it. Report on the table, numbers visible, conclusion drawn before you get there. Sounds like this: The ad performs decently, CTR 2.1%, the issue is landing conversion rate at 0.4%. Look at the product page and checkout. Or: This product has 12% margin. We cannot work on commission at this margin. Either raise the price or we rethink what we promote.

An agency that does not respect you does the opposite. Blames the algorithm, the market, seasonality, last month's iOS update. And keeps billing. That is why the commission model self-filters. You cannot earn commission on nothing. If the product does not sell, there is no revenue for the agency either. A commission agency is motivated to tell you the truth. A retainer agency is motivated to tell you what you want to hear.

I have refused clients because their margin was too small. Not because I did not want to work with them. Because the math did not add up. I told them straight: With this margin, even at ROAS 4, you end up negative after our commission. It does not make sense to take your money. Some left upset. Others raised their price, came back two months later, and now they are profitable.

An exercise you do today

Open two tabs. First one, Meta Ads Manager with CTR and frequency for the last 30 days. Second one, Shopify or GA4 with conversion rate and add-to-cart rate. Compare the numbers with the thresholds above. In 15 minutes you have the diagnosis. You know whether the problem is the ads, the product, the offer, or the checkout.

If the diagnosis points to the product or the offer, do not feel bad. Feel informed. Because the information you have now is worth more than another month of ads thrown at a landing that does not convert.

We do not look for who is to blame. We look for causes. If the product is weak, there is no shame. It is information. If the site is broken, there is no shame. It is information. If the previous agency messed up, there is no shame. It is information. The only shame is receiving the information and not acting.