Almost every week someone who runs an online store asks me the same question: “Where should I spend my ad money, Facebook or Google?” It is a fair question. The budget is small, every euro counts and everyone has an opinion.
I have been selling online since 2005, from MarketOnline to Floria (an online flower shop I started in 2009), and since 2013 I have worked every day with online stores at Retargeting, our marketing automation platform. I have burned money on both channels and made money on both. My short answer is annoyingly simple: it is not “or”, it is “and”, but in the right order and with the right measurement. The long answer is below.
Two channels, two kinds of demand
The real difference is not the platform, it is the person on the other end. On Google, people are searching. They already have a need, they typed it into a box and they are waiting for an answer. Someone typing “black evening dress size 8” is practically holding their card. On Facebook, nobody is searching for anything. They are scrolling through photos of their friends’ kids and memes. You show up in the middle of a conversation nobody invited you to.
Hence the basic rule: Google AdWords captures demand that already exists, Facebook Ads creates new demand. Neither is better. They are simply different tools, like a hammer and a screwdriver.
When AdWords works better
If you sell things people already search for, search is where you need to be first. Car parts, electronics, consumables, gifts for a specific occasion, flowers for a specific day. At Floria I learned early that around March 8 (International Women’s Day, a big flower-giving day in Romania) or Valentine’s Day, people are not in the mood to discover new brands. They search for “flower delivery Bucharest” and buy from the first shop that looks trustworthy and delivers on time.
A few things that make the difference in AdWords for a small store:
- Long, precise keywords, not generic ones. “Shoes” is expensive and brings browsers; “men’s brown leather shoes size 9” is cheap and brings buyers.
- Negative keywords from day one. Otherwise you will pay for “free”, “second hand” and “DIY”.
- Remarketing and RLSA (remarketing lists for search ads), so you can bid harder for people who have already visited your site.
Whatever channel you choose, you need a clean product feed: clear titles, up-to-date prices and stock, good photos. You use it in price comparison sites, in dynamic ads on Facebook and anywhere else you want to show your products with a photo and a price.
When Facebook works better
Facebook shines where the product is visual, new or an impulse buy: fashion, accessories, home decor, gadgets, unusual gifts. Things people did not know they wanted until they saw them.
What I like about Facebook in 2016 is that its targeting has become genuinely useful for commerce:
- Custom Audiences: upload your customer list (emails or phone numbers) and show ads only to them, or exclude them so you stop paying for people who have already bought.
- Lookalike Audiences: Facebook finds people who resemble your best customers. For a store with a few thousand customers, this is often the best cold audience you can build.
- Website Custom Audiences, with the new Facebook pixel: reach again the people who viewed a product or abandoned their cart.
- Dynamic Product Ads: show each person the exact product they looked at, generated automatically from your feed.
Then there is Instagram, where ads opened to all advertisers only a few months ago, from the same interface as Facebook. The audience is younger and more visual, and for now there is little competition for the ad space. If you sell fashion or beauty, I would test it now, while it is still cheap.
Retargeting is where they meet
I admit I am biased here, because this is what I do all day. But the industry numbers have said the same thing for years: the vast majority of visitors to an online store leave without buying. If you pay to bring them in and then let them go without ever reaching them again, you bought traffic just to lose it.
Retargeting is the first place I put money, before any acquisition campaign. On Google (remarketing on the display network and in search) as well as on Facebook. Plus email, which many people forget, even though it is the cheapest channel of all. My order of priorities for a small store: first win back the people who have already been to my site, then capture search demand, and only then create new demand.
How to measure without lying to yourself
- Put the conversion tag on the order confirmation page, with the real order value. Without it you are not doing marketing, you are making donations.
- Turn on e-commerce tracking in Google Analytics and tag every link from Facebook and from your newsletter with UTM parameters. Otherwise half your traffic shows up as “direct” and you have no idea where it came from.
- Watch out for attribution. Google Analytics gives all the credit to the last click by default. Facebook, on the other hand, claims conversions from people who merely saw an ad in the last day or clicked one in the last 28 days. Add the two reports together and they will tell you that you sold more than you actually did.
- Look at assisted conversions in Analytics too. Facebook rarely closes the sale, but it often starts it.
- Measure cost per order and margin, not cost per click. A cheap click that never buys is the most expensive click there is.
My favourite test, simple and honest: switch off one channel for two weeks during a normal period and see what happens to total sales.
How I would split the budget
There is no magic formula, but if you ask me what I would do with a small or mid-sized store starting from scratch, I would begin roughly like this: a fixed share for retargeting on both networks, the bulk of the budget in AdWords search, where demand exists and is easy to measure, and a smaller but steady share for Facebook tests with lookalike audiences and fresh creative. After a month or two, move the money towards whatever brings profitable orders, not whatever looks good in the reports.
If nobody searches for your product on Google, flip the proportions. If it is a pure search product, like spare parts, Facebook may stay a retargeting-only channel. What matters is deciding based on your own data, not on success stories from American blogs.
The mistakes I see most often
- Using the “Boost post” button as a marketing strategy. It brings likes, rarely sales.
- Sending ad traffic to the homepage instead of the product page shown in the ad.
- A site that is slow or awkward on mobile. More and more Facebook traffic comes from phones, and if the page loads slowly you are paying for people who leave before they see the product.
- Drawing conclusions after three days and a hundred euros spent.
- No real offer. No amount of targeting will save an overpriced product, slow delivery or a site that does not inspire trust.
In short
Google brings you the people who are looking for you. Facebook brings you the people who don’t yet know they are looking for you. Retargeting brings both of them back. A healthy online store in 2016 needs all three, but not all on the same day and not with the same budget. Start with what is easiest to measure, learn from the numbers, and only then scale.
Where have you seen better results for your store, on Facebook or on AdWords? And which mistake cost you the most? I would love to hear.