Stay informed with our latest updates!

Excitement and valuable insights delivered straight to your inbox.

Oops! Something went wrong while submitting the form.
arrow to left
February 7, 2023

The State of E-commerce Marketing

Maya Caspi
Head of Product Marketing

New research shows: eCommerce marketing teams are losing an average of 17% of their budget on ineffective ads. 

A beginning of a new year is always a great time to ask ourselves whether the things we do are still working for us and what we can do to improve.

So to help marketers make the most of the new year and their shiny new marketing budgets, we set out to discover how well different ads perform over time and how much of the budget is wasted on ineffective ads.

Over the past month, we conducted an extensive analysis of more than 50,000 ads of high-spending eCom brands on their main performance channels - Google and Facebook.

And the results?  Eye-opening, to say the least: It turns out 29% of ad groups or creatives had a 20% lower return on investment than the target LTV.

Did you know that you may be losing a significant portion of your budget on ineffective ads? 

As we all know, in 2023, profitability is the new growth. With the state of the market dictating this shift towards profitability, most savvy marketers are already focusing on LTV as a critical factor in driving long-term success for their company. 

But are our marketing efforts really supporting our company's long-term profitability strategy?  

As we mentioned, to find this out, we first did a deep dive into over 50,000 Google and FB ads from high-spending eCom brands.  

The goal of our analysis was twofold:

  1. Evaluate the efficiency of these ads (how much revenue they yield, did they bring high or low-value users, did they meet payback expectations?)

      2.    Understand how much of the average marketing budget is wasted on ineffective ads

Initially, our analysis across all channels and customers showed that both 7-day and 180-day performance generally met targets, on average. However, when drilling down to the ad group or creative/keyword level, our analysis revealed this to be misleading, as a significant portion of the budget was being wasted on ineffective ads: 

  • 29% of ad groups or creatives had a 20% (or more) lower return on investment than the target LTV
  • This lost revenue amounts to 17% or more of the marketing budget on average.

Our analysis also revealed that underperforming ads tend to attract "bad"customers over time and are consistent about that, incurring more "bad" spend as time goes on.

Let's put this in real-life context:

Let's say you're running a Facebook campaign. You are looking to achieve 150% ROAS during the first week, meaning you’re profitable on your marketing investment (gross). You know that, on average, 32% of customers will purchase again in the next 6 months, making you net profitable, with an average ROAS of 230% and an average LTV of $230.

Based on this, you set your target customer acquisition cost (CAC) for Facebook at $100.

Looking back at campaigns you ran 6 months ago, you see that your 7-day ROAS was indeed $150 on average, and your CAC was $100 on average, resulting in a return on ad spend (ROAS) of 150%. However, you notice that campaign A had an eventual 6-month LTV of $190 (resulting in 190% ROAS), while campaign B’s 6-month LTV was $260 (resulting in 260%.ROAS) This lower LTV is consistent across different weekly cohorts and seems to be caused by low retention rates among the users attracted by the campaign's creative.

Had you been able to identify this issue sooner, you could have reallocated the budget from Campaign campaign B. However, by the time you realized the issue, both Campaign A and B had already run their course, and it was too late to make changes.

On average, you lose 17% of your budget to campaigns like Campaign A, which have low LTV but are difficult to identify until it's too late to make adjustments. By using a good LTV prediction model, you can react more quickly and avoid these losses.

The takeaway: Accurate as possible predictions can save you a lot of time and money

It can take months to fully understand an ad's LTV. Most companies, especially these days, can't really afford to wait that long. This means it is imperative to identify underperforming ads as early as possible in order to stop the bleeding. 

OK, so what should marketers do to optimize for high-LTV users and drive long-term success for your company?  

Drumroll please… 🥁🥁🥁 It's 2023 resolutions time: 

Here are a few recommendations based on our research findings:

1. Use predictive models /data to understand future performance earlier. In our analysis, we used actual data, but you should use predictive data (because you can't afford to wait 9 months for results) to understand future performance early on

One way to do this is by building an in-house LTV model or working with a prediction partner.

2. Once underperforming ads have been identified, there are several steps that marketers can take to address the issue:

  • Turn off inefficient campaigns or the bottom-performing percentage of inefficient campaigns
  • Group together creatives by predicted ROAS to create new ad groups and lower the bid according to the expected ROAS
  • Re-allocate the budget to double down on ads or keywords that generate high LTV

3. Lastly, consider incorporating LTV at different maturities (30, 45, and 60 days) into your measurement process. This will give you a more holistic view of your ad groups' performance over time and allow you to make more informed decisions about your acquisition efforts.

Let's sign off for now with this - shifting your campaigns' focus to LTV is not just a strategy, It's the new normal. The best thing you can do this year is to make sure the efforts you put in, will eventually be worth your while. A great strategy, by the way, in marketing and life in general. 


stay updated

join our newsletter

Request Sent. We'll be in touch!
Something went wrong. Please try again.

Continue Reading

No items found.


Stop Retrospecting. Start Futurespecting.

Request Sent. We'll be in touch!
Something went wrong. Please try again.