1. Use Flexible Thresholds:
Instead of relying on absolute values, consider using flexible thresholds like Target ACOS (Advertising Cost of Sale) and Statistically Significant Clicks. This approach ensures that the ruleset adapts to the unique characteristics of each campaign based on their respective thresholds. By doing this, you can use the same ruleset across your account without the need to create separate rulesets for each campaign. This simplifies your management process and helps maintain consistency.
2. Implement "Days since Last Change" on Rule Level:
Use the "Days since last change" parameter at the rule level. This feature acts as an additional safeguard against rapid, compounding changes. If you set a rule to have a "Days since last change" greater than 2, it means that if that rule made a change to a target today, it will wait for 2 days before making another change to the same target. This approach prevents excessive bid adjustments on a single target in a short timeframe and allows the system to observe target performance over a reasonable period before making further changes. It enables you to configure rulesets to run daily without concerns about bids compounding unnecessarily.
3. Carefully Sequence Rules within the Ruleset:
It's crucial to pay close attention to the order and sequencing of rules within a ruleset.
Let's consider an example to illustrate this point:
Example: The current ACOS of a target is 90%, while your target ACOS is 20%. You have now set up a bid management ruleset with the following rules:
Rule 1: If ACOS > Target ACOS * 2 (Current ACOS is greater than twice the target ACOS), then decrease the bid by 20%.
Rule 2: If ACOS > Target ACOS * 4 (Current ACOS is greater than four times the target ACOS), then decrease the bid by 30%.
Now let's understand how the changes will be different as we change the sequence of the rules.
Rule Sequence 1 (Rule 1 is above Rule 2):
Rule 1 checks if the ACOS is greater than Target ACOS * 2 (90% > 40%, as 2 * 20%).
Since the condition in Rule 1 is met, the system executes its action, reducing the bid by 20%. No further rules are checked.
Rule Sequence 2 (Rule 2 is above Rule 1):
Rule 2 checks if the ACOS is greater than Target ACOS * 4 (90% > 80%, as 4 * 20%).
The condition in Rule 2 is met because 90% is greater than 80%, and the system executes its action, reducing the bid by 30%.
In this example, the sequence of rules within the ruleset impacts the result. If you prioritize Rule 2 (ACOS > Target ACOS * 4), the bid would have been decreased by 30%, which is the intended action here. However, if Rule 1 (ACOS > Target ACOS * 2) is sequenced first, the bid reduction is 20% as the criteria is met, and the second rule will not be executed.
4. Use the "Clicks" threshold along with ACOS in rulesets:
It is recommended to consider both the Clicks and the ACOS when setting up automation through rulesets. This combination ensures that any adjustments made by the automation are after analyzing a substantial amount of data over a reasonable period of time, leading to more informed and reliable changes.
Let us take an example to understand it better:
Let's consider a scenario where the focus of automation is solely on monitoring ACOS to modify bids. Imagine a rule that aims to decrease bids for keywords with ACOS greater than the Target ACOS. This might seem reasonable, but there's a key issue with this condition.
In cases involving keywords with 0 orders and a low number of clicks, typically 2-3 clicks, the ACOS is calculated as infinity. This occurs because ACOS equals Spend/Sales, and keywords with clicks greater than 0 will have some spend but no sales, resulting in an infinite ACOS value. If a rule relies solely on the ACOS condition, it may prematurely reduce the bid without waiting for a sufficient number of clicks to make a statistically reliable assessment.
Therefore, incorporating Clicks along with ACOS allows data to accumulate over time, making your ruleset more adaptive and accurate, helping you avoid potential pitfalls, and maximizing the effectiveness of your automation strategy.
How to determine the Clicks threshold?
There are two methods of determining the Clicks threshold:
Calculating the Clicks threshold manually:
Analyze Conversion Rate (CVR): Begin by assessing the Conversion Rate (CVR) for your account or for the specific category.
For instance, if your account's Conversion Rate is 8%, it implies that roughly 13 clicks are needed to generate 1 order.Optimal Threshold Values: It's advisable to use a threshold value that is 5 times this calculated value for setting up the threshold in the Pausing keywords rule, and use a value that is 2 times the calculated value for rules related to bid increases or decreases initially. Over time, you can fine-tune these thresholds based on your account's specific needs and the frequency of changes happening to optimize your advertising strategy.
For example, with an 8% Conversion Rate, the suggested threshold for the Pausing keywords rule would be 13 x 5 = 65 clicks, and for bid-related rules, it would be 13 x 2 = 26 clicks.Inputting the Manually calculated Clicks into rules: To input the manually calculated Clicks threshold into the ruleset, add a condition for Clicks in the Criteria section and set the value to "Absolute Value" and input the calculated number (e.g., 65 clicks for pausing keywords).
Using the Statistically Significant Clicks (SSC) calculated by Adbrew:
Statistically Significant Clicks can be defined as the minimum number of clicks required to generate at least 1 order. It is calculated by the system on the basis of the overall Conversion Rate (CVR) of the adgroup. Statistically Significant Clicks can be found under the Values drop-down while setting the "Clicks" threshold in the rulesets.
Note: The exact value of SSC for any product or ad group cannot be viewed as it is dynamically calculated by the system continuously for each and every individual product or ad group.
5. Use MAB cautiously and make sure to connect your seller/vendor account to Adbrew:
MAB stands for "Maximum Affordable Bid," which is the highest bid amount for a keyword required to maintain the Target ACOS. It is used to define the Action to be performed by the rule. The formula used by the system to calculate MAB is:
Max Affordable Bid (MAB) = Average Price X Target ACOS X Conversion Rate X Current Bid/ Cost per click (CPC)
It's essential to use MAB cautiously, as it can aggressively increase bids and may lead to higher ad spend. It's important to note that for the system to successfully calculate MAB for new keywords with no historical data, you should ensure the system has the necessary historical data and fallback values by connecting your seller/vendor account to Adbrew. To learn how to connect our seller/vendor account to Adbrew, click here.
6. Make sure to exclude Brand and Hero terms from negative automation:
Carefully review and fine-tune your negative automation to ensure it doesn't unintentionally negate important branded or hero keywords in your campaigns. To achieve this, consider adding specific conditions in your negative targeting ruleset, such as "Search term does not contain Brand term for Account/Product" and "Search term does not contain Hero term for Account/Product," as shown in the below screenshot. These conditions will help exclude and safeguard your brand or hero terms from negative automation.
7. Use Min/Max thresholds in the bid management ruleset to make sure your bid remains within a certain pre-defined limit:
Consider configuring Min/Max bid thresholds in your bid management rulesets when you want to establish the minimum and maximum bid limits, ensuring that automation doesn't increase/decrease your bid beyond a certain point. This means that bids will never be increased beyond the maximum threshold, and they will not drop below the minimum threshold. Using these thresholds allows you to have more precise control over your bid management within defined limits. In order to find the option to set the min/max threshold:
Navigate to the specific bid increase or decrease rule you wish to modify and under the "Actions" section for that rule, click on "Show advanced settings."
In the advanced settings, you will find the option to input minimum and maximum values.
NOTE: Ensure that your Max Bid threshold (the maximum value) is set higher than your current bid. Otherwise, the automation will reduce your current bid to match the maximum value you've set.
8. Make sure to keep the lookback period of the ruleset to at least 7 to 14 days:
It is recommended to set up your automation to make changes based on a minimum of 7-14 days' worth of data. This approach ensures a more accurate evaluation of your campaign's performance and enables the automation to avoid making impulsive decisions influenced by short-term fluctuations.