I stopped advertising and nothing happened

If you found this post interesting, let me know on Twitter: https://twitter.com/davidjames

Background

At TotalCarCheck.co.uk, between 2017 and 2021 (when the company was mostly just me) I was paying for in-search adverts across multiple search engines, I was commissioning custom made adverts to place in online used car market places. I was spending time optimising the 'bid' in automated advertising system, trying to get the best bang for buck.

Why was I doing it? Everyone else seemed to be doing it.

This was a terrible reason to do it.

In the end I decided the only real way to tell if what we were doing was working, was to turn it all off, for several months to see if there was a measurable difference. In my mind, worse case scenario, I could see after a few weeks if it was a terrible idea and turn the ads back on. We never did turn them back on.

I'd always built Total Car Check with SEO being at the top of my mind, if we could outrank our competitors when users searched for the most popular terms, we could do OK, and we were, we were doing great, but somewhere at the back of my mind I felt that we had to use paid-for advertising to supplement this organic traffic. I was wrong.

Lessons Learned

The top 4 things I learned were:

0) Start with a well designed SEO friendly site. If you don't know where to start, just use:  Googles own PageSpeed tool, there is so much low hanging fruit. Concentrate your efforts here. A fast, mobile friendly site ranks well, this will reduce your costs of acquisition in the long term. If you do not do this as a first step, you will become addicted to paid traffic, and like any recovering addict, the road back to cold turkey town is long and hard. Mark my words.

1) Online adverting isn't all evil. You just have to know when to start and stop. Using paid for advertising when you are starting out (i.e. low number of users), and specifically things like Google AdWords is actually fine, you have a low number of users, you need to increase awareness of your product. If you can balance it so you are breaking even at worst, that's ok. People will get to know your product, you will get users, which leads to feedback, which is important when growing. Also, I have MASSIVE CONFIRMATION BIAS. This was my experience in my niche, there are millions of dissimilar products and services. For example, I know the physical product market works very differently to the used vehicle provenance market. Apply salt as required. If you find, and can measure that for every £1 in ad spend means £10 in sales, good luck to you. However, check you're not paying for your own traffic, see below.

2) At some point, you are going to be paying for traffic that was coming to the website anyway, and maybe in a "wow, this is crazy" sort of way. Over 85% of our organic search traffic comes from users typing in our brand name - Total Car Check. If they see an advert at the top of the organic search results, for the user, its just a shortcut to get to the site, especially since Google started putting organic search results below the fold in many search results, except it would cost us money to buy a user who was going to find us anyway. There has been a decent amount of research on this topic, I listen to a Freakonomics podcast on the topic, which I highly recommend, and a Harvard Business Review article that describes how eBay turned off millions of pounds of advertising and found:

... overall, there was no appreciable decline in sales of eBay listings in the part of the country where Google ad purchases were shut off.

Some multi-national organisations have turned off hundreds of millions of pounds of advertising, and seen, no discernible change in sales or conversion.

3) The stats you see when you buy adverts suggest they are always working, because, they are, but what you are seeing is a form of a self fulfilling prophecy, people that are coming to see you, are going to spend money with you. This makes it appear as though your ROI is amazing, when in fact, you just increased your cost of acquiring a customer by a factor of infinity. (i.e. the cost to acquire a customer who was coming to buy anyway should be zero.)

Conversion value per cost measures return on investment. It’s the conversion value divided by the total cost of all advert interactions. Simply put it measures how many £'s your customers spend when you spend £1 on advertising.

The chart above shows our average conversion per cost over the lifetime of our adverts. The average was 3.25, meaning we were being told for every £1 we spend, a user spent £3.25. Sounds amazing right? Just spend more money on adverts, surely. Nope. We turned off all adverts. Sales increased.

We also tried other forms of advertising. Warning bells starting ringing in my ears, when on one particular occasion we spent £5,000 (+ vat, no less), for a large deal with a popular online car market, created optimised images and copy, designed for desktop and mobile and ultimately, saw, wait for it.... TWO sales. Again, your mileage will vary on your market segment, but be aware that in the direct to consumer market, instant conversions are hard. There is a solid case to be made about increasing your brand awareness via this sort of advertising, but be true to yourself if this is what you are telling yourself.

Summary

I guess my one big take away from all of this is that not doing something needs to be tried as actively as doing things, or put correctly, subtracting from instead of adding to solutions needs to be considered more openly, there was an interesting tweet on this topic.

For the New Year, subtract something to move forward. Breakthroughs happen when we remove stuff. After all, Han Solo is just a cowboy without a hat, as @simonstalenhag illustrated. But we default to adding things when solving problems, often just creating more problems & burdens. pic.twitter.com/TqJUoAnr7z

— Ethan Mollick (@emollick) December 31, 2021