I was at a conference recently and one of the speakers said, “Smart Shopping is the devil.” I admit I shared that sentiment when Smart everything first rolled out (we cautiously test automation to ensure it produces an ROI on a client-by-client basis). However, after testing Smart Shopping across a number of accounts, I changed my attitude.
Before I sing the praises of Smart Shopping, let’s review some of the points that support the opinion that Smart Shopping is the devil:
- No placements report and no negative placements. This is the biggest risk when running Smart Shopping. Without the ability to see and control your placements, your brand could be appearing in and alongside of content that does not reflect the values of your company. *Update 6/2019: @PPCKirk shared this post on pulling a report on placements. Still can’t exclude any placements, but at least you can monitor the placements and pull the plug on Smart Shopping if you don’t like what you see.
- No search term report and no negative keywords. The search query report is a GOLD MINE! Both for keywords to target and to exclude.
- No learning. Without all of this data, there’s no way to leverage the wins and learn from these campaigns. I’d love to see what queries are driving these sales and add them to my Search ( keyword / text ads ) campaigns. I’d test them on Bing, too. It would really be great to share some queries with my SEO colleagues to consider for content opportunities. But Smart Shopping reveals nothing.
- No attribution. Since Smart Shopping combines search and display, channel attribution is impossible. Google hopes that you’re too busy counting your money to care.
Case Studies
Despite this lack of transparency, it was still worth a test. We’ve rolled out Smart Shopping in a handful of accounts, and have shown great success in some, and big fat failures in others.
Auto Parts
Manual Shopping Campaigns: typical monthly ROAS between 7-8x.
Smart Shopping Campaigns:
- Month 1: 9x
- Month 2: 10x
- Month 3: 12x
- Month 4: 13x
- Month 5: 14x
Not only has the ROAS improved, but we’ve been able to scale it. With the stronger margins, the client has increased ad spend by 50% and growing.
Home Decor
Manual Shopping Campaigns: typical monthly ROAS between 2-3x
Smart Shopping Campaigns:
- Month 1: 3.5x
- Month 2: 4x
- Month 3: 4.3x
- Month 4: 4.4x
Bedding
Manual Shopping Campaigns: typical monthly ROAS 2-3x
Smart Shopping Campaigns:
- Month 1: 3.9x
- Month 2: 4.3x
- Month 3: 3.12x
- Month 4: 2x
Electrical Supplies
Manual Shopping Campaigns: typical monthly ROAS between 8-10x
Smart Shopping Campaigns:
- Month 1: 5x
- Month 2: 2x
Keys to Success
Based on the accounts where Smart Shopping has been successful, these are the factors I believe to be the keys to success:
- Strong manual performance. To be eligible for Smart Shopping, Google’s minimum is 20 conversions in the past 45 days from Shopping ads. But just because your account meets those minimums does not necessarily mean Smart Shopping is going to work. You need to have a history of conversions with manual campaigns for the machines to learn from.
- Strong audience lists. Smart Shopping eligibility requires you to have the remarketing tag on the site. But again, just because you put the remarketing code on the site doesn’t mean Smart Shopping is going to work. The accounts that have shown the best success with Smart Shopping have a history of audience data.
- Conversions tracked with the tag, not imported transactions from Analytics. Despite being told by Google that the automation is looking at “the conversion column” regardless of the type of conversion, we’ve tested this in accounts that were importing transactions from Analytics, and Smart Shopping results were weak. Once we switched to the conversion tag, Smart Shopping started to shine.
If your account meets the above criteria, test out Smart Shopping. Sure, maybe it is the devil; but you can’t argue with results.
Not familiar with Smart Shopping? Start here.