Podcast: Download (Duration: 56:25 — 64.9MB)
Start YouTube ads with your remarketing lists rather than cold audiences, because that is where the predictable wins are. Brett Curry compares top-of-funnel video to a half-court shot at the buzzer, and remarketing to a layup.
Brett runs OMG Commerce, a Google Premier Partner agency that has handled pay-per-click for over 125 companies. He wrote the Ultimate Guide to Google Shopping published by Shopify.
Shoppable TrueView is the format he is most excited about. Your video ad runs with product listing ads beside it, pulling their images, prices, and titles from the same feed that powers Google Shopping.
This episode covers which remarketing audiences convert best, what view-through rates to expect, how to hook viewers in five seconds, and the attribution problem that makes YouTube ROI harder to read.
A note on timing: this is a 2017 conversation. Google’s ad formats, naming, and attribution options have changed substantially since.
Get My Free Mini Course On How To Start A Successful Ecommerce Store
If you are interested in starting an ecommerce business, I put together a comprehensive package of resources that will help you launch your own online store from complete scratch. Be sure to grab it before you leave!
Table of Contents
Key takeaways
- Start with remarketing lists, since cold-traffic video rarely produces direct conversions.
- Audience quality ranks from all site visitors to product page viewers to cart abandoners, each converting better.
- The bought-X-but-not-Y list is small and among the strongest performers.
- View-through rates run 9% to 12% on the low end and 20% to 30% for compelling, well-targeted video.
- Cost per view typically runs nine to twelve cents, occasionally as low as a penny.
- Hook viewers within five seconds, because that is when the skip button appears.
- Production value matters far less than clarity, and phone footage works.
- Combine direct and view-through conversions when calculating whether you hit your target.
What is shoppable TrueView?
A YouTube video ad with product listing ads displayed alongside it. Those product ads pull from the same feed that powers Google Shopping, showing image, price, and title.
TrueView means you pay only when someone watches. The skip button appears after five seconds, and the advertiser is charged only if the viewer watches 30 seconds, the full video, or interacts with it.
That self-selection is the advantage over forced-view formats. Six and eight second bumpers suit large brand advertisers who only want impressions.
The placement encourages shopping directly. Products sit beneath the video on mobile and beside or above it on desktop.
Which YouTube audiences should you target first?
Remarketing lists, ordered from least to most likely to convert. All site visitors, then product page viewers, then cart abandoners.
Recency matters within each list. Segment by 180, 90, and 30 days, with more recent visitors converting better.
Cart abandoners perform best in Brett’s accounts. One client came in at roughly a third of their cost-per-acquisition target on that audience.
Fourteen or seven day windows work best for carts. Tighter recency concentrates intent.
All-site-visitor lists are inconsistent. They work well for some clients and poorly for others.
What is a bought X but not Y audience?
A custom list of people who purchased one product from you and never bought another. It is small and converts unusually well.
Brett’s example is a skin care brand. They built a list of buyers who had not purchased one specific moisturizer, then ran a video featuring exactly that product.
The product listing ads beside it reinforce the message. The customer already trusts you, and you are showing them a gap in what they own.
Volume is limited by list size. Total conversions stay modest even at strong efficiency.
What view-through rate should you expect on YouTube?
Between 9% and 12% at the low end, and 20% to 30% when the video is compelling and tightly targeted. View-through rate measures the percentage completing 30 seconds or the full video.
Below 10% points to one of two problems. Either the video fails to engage, or the targeting is wrong.
Targeting problems are unlikely on a remarketing list. Check whether a general audience got mixed in.
Longer videos have lower rates for a mechanical reason. A four-minute video still measures against the 30-second threshold, so an 11.78% rate on one is not comparable to a short ad.
Where people drop off matters more than the headline number. Viewers leaving when your offer appears may be clicking through, which is what you want.
What does a YouTube ad cost?
Nine to twelve cents per view is the typical bid range, and Brett has seen it as low as a penny. You are charged your cost-per-view rate rather than a separate click charge.
Clicks are free in that sense. Someone clicking your overlay costs you the same view charge.
Google’s calculated cost-per-click looks alarming. Dividing total spend by clicks produces a few dollars, which compares badly to search until you understand the mechanics.
How do you make a YouTube ad that people watch?
Hook them in the first five seconds, because that is when they can skip. Brett rejected a client video that opened with six seconds of logo animation.
For a grill company, that meant leading with food. Ribs on a grill with steam rising rather than branding.
Lead with your strongest benefit. What the product does for the viewer, whether saving time, saving money, or improving an outcome.
Show rather than list. Video’s advantage is demonstration, and flashing bullet points wastes it.
Humor works when it fits your brand. A widely shared bathroom spray ad opens with a line startling enough that viewers stay.
Do YouTube ads need professional production?
No. Brett explicitly warns against spending $5,000 on production to test the channel.
Phone footage works when it is clean and to the point. It should avoid looking sloppy without needing a high finish.
Customer testimonials are a strong low-cost format. Someone describing their results after 30 days speaks directly to the audience.
Treat it as a minimum viable product. Get something that speaks to customers, highlights the product, and carries a call to action, affordably.
How do you measure YouTube ad performance?
Combine direct conversions with view-through conversions before judging against your target. Someone may watch without clicking and buy later.
Default attribution is last click. Direct conversions only count people who clicked through and bought immediately.
View-through conversions capture the rest. Someone who watched, did not click, and purchased within the attribution window.
The overlap creates double counting. A viewer who later clicks a text ad gets counted both as a view-through conversion for the video and a last-click conversion for the search ad.
Brett’s guidance is to aim for less wrong rather than exact. If click-through conversions are only barely hitting your target, some of those view-through conversions are probably attributed elsewhere too.
Cross-device tracking is improving and imperfect. It generally requires the user to be signed in across devices.
Should you use CPA or return on ad spend?
Return on ad spend for most catalogs, and about 80% of Brett’s clients use it. Revenue divided by cost handles varied price points.
CPA works when order value is consistent. A client selling a narrow product range can use a fixed cost-per-acquisition target.
Wide catalogs break CPA. A client with over 100,000 products cannot treat a $10 sale and a $1,500 sale as equivalent.
Set the target from your margin. At 25% margin, breaking even on the first order requires a 4x return on ad spend.
Repeat purchase behavior can justify a looser target. Businesses expecting several orders in the first year can afford more upfront.
Should you run YouTube ads to cold traffic?
Not until remarketing is working. Brett is candid that his agency is still experimenting here with clients who accept it as a test.
The most promising cold approach combines in-market audiences with keyword targeting. In-market means people whose behavior signals they are shopping a category.
Keyword targeting narrows it further. Layering recent YouTube searches on top of in-market status produces a qualified audience.
Similar audiences are worth testing too. Built from your converter list, comparable to a lookalike audience.
Cold traffic wants a different landing page. Send them somewhere informational and capture contact details rather than pushing straight for a sale.
What minimum budget do YouTube ads need?
A remarketing list of at least 1,000 people, and ideally 5,000 or more. Below that, visibility is too limited to matter.
Roughly $500 a month is a common test budget. That buys enough views to produce meaningful data.
The real cost is the video. Creating the asset is the expense worth planning around.
Small stores may not spend their budget. With limited traffic and remarketing-only targeting, you may not reach your cap.
Frequency capping is worth setting. Around three views per week is a reasonable starting point.
How do you fix an underperforming Google Shopping campaign?
Check the feed, then the campaign structure, then the bids. The feed is where most problems live.
Titles and descriptions are the common failure. Weak titles without keywords give Google too little to work with.
Category mismatches confuse Google. An incorrect product category throws off relevance.
Disapproved products are frequently overlooked. Many accounts have items silently rejected.
Structure by funnel position. Broad queries like a bare category name should surface your top converters at higher bids.
Bid at the SKU level. Separating winners from losers lets you feed the products that convert and starve the ones that do not.
Products with zero impressions need diagnosis. Low search volume, insufficient title detail, an incorrect out-of-stock flag, or bids too low to compete.
Should you focus on SEO or paid search first?
Paid, though do both. Paid gives you control and immediate data, while SEO takes many months.
Zero traffic hurts your rankings. Google appears hesitant to rank a site nobody visits, so paid traffic helps the organic effort.
Paid data improves your SEO. Headlines and copy that convert in ads can inform title tags and meta descriptions.
Category pages deserve attention. Authority built there passes through the site hierarchy, though new sites usually rank product pages first on long-tail queries.
Frequently asked questions
What is shoppable TrueView?
A YouTube video ad with product listing ads displayed alongside it, pulling images, prices, and titles from the same feed that powers Google Shopping.
What audiences work best for YouTube ads?
Remarketing lists, ordered from all site visitors to product page viewers to cart abandoners. Cart abandoners and bought-X-but-not-Y lists convert best.
What is a good view-through rate on YouTube?
Nine to 12% on the low end and 20% to 30% for compelling, well-targeted video. Below 10% suggests either a weak video or a targeting problem.
How much do YouTube ads cost?
Nine to twelve cents per view is typical, occasionally as low as a penny. You pay your view rate rather than a separate charge for clicks.
How long do you have to hook a YouTube viewer?
Five seconds, which is when the skip button appears. Lead with your strongest benefit and show the product rather than listing features.
Do YouTube ads need professional video?
No. Phone footage and customer testimonials work provided the video is clean and clear. Avoid large production budgets when testing the channel.
What budget do you need for YouTube ads?
A remarketing list of at least 1,000 people and roughly $500 a month for a meaningful test. The video itself is the main cost.
Should you run YouTube ads to cold audiences?
Only after remarketing is working. Cold video rarely produces direct conversions, and the best approach combines in-market audiences with keyword targeting.
