4 ms·
A lot of people use Google to find things. A lot of those people click the first search result that matches what they're looking for (low cost, certain feature,
by jlbnjmn 6y ago
A lot of people use Google to find things. A lot of those people click the first search result that matches what they're looking for (low cost, certain feature, prestige, there's no right answer, just right answer for them).
Some of those clicks are for products and services with short buying/lead gen cycles. Once they've clicked, many are only going back if they don't like what they find.
The value of search ads is being the link they clicked. You can buy that click. You can literally insert your website into a demand/supply match. If you don't mess it up after they click, those clicks become transactions. In dollars.
Oil changes are a great example. Some of the people who type "oil change near me" in maps don't really care about the brand. They click the top result, schedule an appointment, and get their oil changed.
If the click was $3 and it took 10 people/bots clicking to set one appointment, that's $30. Now let's say the oil change is $39 so profit is ($39 - oil $7, - filter $3, - labor $5, - ads $30) = -$6 before the cost of skilled advertiser.
Are those ads profitable? It depends. If you can upsell 1/10 on replacing a cabin air filter for 10x cost, then yes.
Every other upsell becomes profit.
Also, at a car dealership you can sell 3-5% a car while they're there for a nice $1700-$3200 profit, which nicely covers the cost of the skilled advertiser.
- qppo 6y agoA huge point of the article is that valuing search ads like that is highly susceptible to selection bias. My point was that the value of an ad is to displace other ads, in other words the measurement you look for isn't how many people click through to your site for an oil change but how many do so instead of the next guy.