Smart Bidding is like having a mathematician on your team—good with historical data but blind to future events.
While it excels at optimizing bids based on past performance, it can’t predict that Black Friday is coming or that your summer sale will triple conversion rates.
For us Google Ads managers, this creates a critical challenge: how do you guide an AI-powered system through highs and lows in demand?
That’s what we’re going to dive into today.
The Challenge with Smart Bidding
I love Smart Bidding. It’s an incredible tool that has made Google Ads management much easier. But there’s one limitation you need to understand: Smart Bidding is reactive, not predictive.
It can’t predict performance. It doesn’t know about current events. All it can do is review historical performance and adjust bids up or down based on this data.
And as perhaps the best-known fact about performance goes: historical performance does not equal future performance.
This creates three situations where you need to step in:
- Peak Season: Think summer for swimwear
- Promotions: Like big weekend sales
- Black Friday: Where everything works differently
Let’s look at why these three need different approaches.
Why Each Situation is Different
Peak season is slow and steady. Customers don’t stop buying the moment peak season ends, so you have time to adjust. That’s why quick fixes like temporary adjustments to Smart Bidding often hurt more than they help.
Promotions are quick sprints. You need to bid up fast to catch all the demand, then bid down quickly afterward to avoid wasting money. Plus, you have to prevent Smart Bidding from relying on that high-performance data for too long.
Black Friday is unique because everyone’s doing it. Running 20% off on a random weekend? Great—bid 10-40% higher, and you’ll do well. But during Black Friday, everyone’s bidding higher, and everyone’s running promotions.
Managing Peak Season
Getting Ready
Don’t use temporary adjustments during peak season—they’re short-lived and will hurt when they expire. Instead, focus on setting your Smart Bidding targets correctly:
- Lower ROAS by 10-20% to help Smart Bidding find new converting segments.
- Start this 2-4 weeks before peak season because:
- You don’t know exactly when the peak will hit.
- Smart Bidding needs time to identify those new segments.
- Increase ROAS again after your first strong week in peak season.
During Peak Season
Group your campaigns by performance. If your main target is 600% ROAS, here’s how to do it:
- High ROAS: 800% target
- Don’t waste money chasing 100% impression share.
- 95% is plenty.
- Main ROAS: Your standard target
- This is where most of your spend should be (50-80%).
- Low ROAS: 500% target (or lower)
- The goal is to capture demand from campaigns that aren’t hitting volume at your ideal ROAS target.
- Sometimes, you just need to lower your ROAS slightly to capture huge demand.
Don’t overdo it. I see too many people creating 10+ portfolio bid strategies and splitting their data too thin. Keep most campaigns in your main group. Only move the clear winners and losers out.
After Peak Season
Usually, you don’t need to do anything. Let Smart Bidding adjust as performance naturally drops.
Lowering your max CPCs can help steer Smart Bidding toward reducing bids.
But watch out for sudden drops—like when you’re selling swimwear and the weather turns bad during your last peak month. In such cases, you might need to step in with stronger adjustments.
Managing Promotions
Getting Ready
Lowering your ROAS target before a short promotion can hurt more than it helps. You don’t have time to lower and then rebuild the ROAS target during a 3-7 day promotion.
You have two ways to approach it:
- Spend high before the sale (frontload ad spend):
- Build interest early.
- Convert them during the sale.
- Run a higher ROAS during the sale.
- Keep spend normal, bid high during the sale:
- Wait for the sale to start.
- Bid aggressively when conversion rates jump.
There isn’t a one-size-fits-all solution, but here are some thoughts on when each scenario works best:
Spend high before the sale if you sell:
- Higher AOV products
- Items people research extensively
- Example: A $5,000 mattress won’t sell just because it’s on sale this weekend.
Keep spend normal if you sell:
- Lower AOV products
- Run a multi-brand store where you capture existing brand demand anyway
- Example: Nike shoes will sell during your sale—you don’t need to create demand.
Changes made:
- Improved sentence flow for better readability.
- Added colons and bullet points for clearer formatting.
- Corrected “one-size fits all” to “one-size-fits-all” for grammatical accuracy.
- Minor rephrasing for conciseness and clarity.
During the Sale
Keep your typical ROAS target. Any changes to your ROAS target won’t react fast enough to bid higher during a short promotional period (unless it lasts more than 7 days).
Keep your normal ROAS target:
- Changing ROAS takes too long to work for short sales.
- Use Seasonal Bid Adjustments to push spend up.
Try increasing bids on fallback campaigns that usually don’t perform well:
- Sales can make poor performers turn profitable.
- Increase your minimum bids on bid strategies to push for higher volume.
After the Sale
The shorter your sale, the faster you need to adjust afterward:
- Use Seasonal Bid Adjustments and Data Exclusions.
- For data exclusions:
- Don’t go beyond 3 days (7 max).
- More than that, and Smart Bidding loses too much data.
Bigger sale = more important to exclude data.
Managing Black Friday / Mother’s Day, etc.
From here, I’ll just write Black Friday, but the same rules apply to big moments throughout the year: Mother’s Day, Father’s Day, Labor Day, etc.
Black Friday is different. Everything you know about normal sales needs to change.
Getting Ready
Use the same approach as normal sales (see above for expensive vs. cheap products). However, because consumers expect promotions to drop for Black Friday, they are much more likely to hold off on buying anything until they see a promotion live.
During Black Friday
Here’s the big difference:
Normal sales: Set higher bids once or twice, and you’re good.
Black Friday: Keep increasing bids because everyone else is. The goal is to constantly review your results (ROAS, revenue volume, blended ROAS) and keep pushing bids higher.
This is VERY different from your normal promotional periods, where not every competitor is continuously bidding higher. During regular sales, a one-time bid increase is usually enough—but not during Black Friday.
What to watch:
- ROAS
- Revenue
- Overall ROAS
- Where your ads show up
Here’s what happens if you don’t:
You increase bids by 20% on Monday. By Thursday, you’re at the bottom of the auction because everyone else kept pushing their bids higher.
After Black Friday
This is when you need to be most aggressive:
Use data exclusion to exclude the biggest days. Google says it’s not necessary, but that assumes:
- What sells during Black Friday also sells after Black Friday. Often, this isn’t true, as the promotions during Black Friday are much bigger. This is especially true for multi-brand stores.
Decrease bids after Black Friday.
Use Seasonal Bid Adjustments to aggressively bid lower. My point of view is that you have everything to lose and nothing to gain in the days after Black Friday.
For some, you move into the Christmas season, which converts almost as well as Black Friday. Even then, I’d recommend starting slow in the days after Cyber Monday ends and gradually ramping up.
Seasonal Bid Adjustments should therefore be aggressive:
- Tuesday/Wednesday: Minus 50%
- Thursday/Friday: Minus 25%
- Saturday/Sunday: Minus 10%
- Then return to normal bidding.
This can—and should—change based on your brand. Some eCommerce brands perform well over weekends, in which case I’d suggest easing off the Seasonal Bid Adjustments sooner to return to normal bidding by the weekend.
The Surprising Metric to Keep an Eye On
One number tells you if your strategy works: CPC.
Click prices, really, Andrew?
Hear me out.
Since ROAS is a delayed metric (you can’t see performance live, and conversion lag complicates things), you need a live metric to track whether your bids are correct during and after promotions like Black Friday.
Here’s how we do it:
- If the normal avg. CPC is $1.30, and you see $2.10 after a sale, it means Smart Bidding is overspending, which will result in a low ROAS.
So, we calculate what the ideal CPC should be before, during, and after a big sale. This becomes the metric we use to set our Seasonal Bid Adjustments.
Your CPC acts like a canary in the coal mine. If it’s too high, something is wrong.
Wrapping Up
Smart Bidding is great at what it does. I love being able to guide it with what we’re good at: seeing what’s coming.
Smart Bidding will beat me at math every time. But it can’t predict what’s coming next like we can.
Success comes from knowing when to let Smart Bidding work and when to step in and help. Use what we covered today to identify which situation you’re in and what to do about it. Remember: we’re not fighting Smart Bidding—we’re helping it handle things it can’t see coming.
