Troubleshoot your bid strategies

If a bid strategy isn't performing as expected, add troubleshooting columns to the bid strategy reporting tables. The columns display data that can help you understand why the bid strategy is performing as it is.

To add troubleshooting columns:

  1. Navigate to an advertiser.

  2. Click Bid strategies in the left navigation pane.

  3. To focus on the performance of specific biddable items, click the name of a bid strategy in the Bid strategies column.

  4. Select a date range that begins on or after the day that you applied the bid strategy to campaigns, ad groups, or biddable items.

  5. Add any of the columns described below.

Reporting columns for troubleshooting bid strategies

Bid strategy column Description

Bid strategy health

Helps you diagnose and fix issues that prevent your bid strategy from performing optimally. Learn more.

% at max bid

% at min bid

The percentage of days within the selected range when biddable items reached their maximum or minimum strategy bids. You may want to take action if the percentage is high, as you might not achieve your goals within the specified bid range. For example, if your keywords reached the max bid a high percentage of days, you may want to raise your max bid.

% cost err at max bid

% cost err at min bid

% pos err at max bid

% pos err at min bid

The percentage of error that can be attributed to biddable items being constrained by their bid limits. To lower these percentages, you may want adjust your max bid or min bid. For example:

  • A value of 30 in the % cost err at max bid columns: 30% of errors are caused by DoubleClick Search wanting to raise a bid to help reach the cost goal, but being prevented from doing so because it already reached the max bid value. To lower this percentage of error, you may want to increase your max bid.
  • A value of 20 in the % pos error at min bid column: 20% of errors are caused by DoubleClick Search wanting to lower a bid to help reach the position goal, but being prevented from doing so because it already reached the min bid value. To lower this percentage of error, you may want to decrease your min bid.
Within these percentages, there may occasionally be factors outside bid constraints that caused the errors.

% of optimal cost

How your actual cost compares to your optimal cost. This is calculated as actual cost / optimal cost:

  • A value of 100% indicates that your actual cost was exactly on target with your bid strategy.
  • A value less than 100% indicates that the bid strategy is over-performing.
  • A value greater than 100% indicates that the bid strategy is under-performing.

% without clicks

% without conv

% without impr

The percentage of days within the selected range when biddable items did not have clicks, conversions, or impressions.

  • If any of these percentages are high, you may want to take a look at the existing biddable items in the bid strategy and add more useful ones.
  • If the % without clicks is very high, it might be helpful to raise your max bid. If it’s a position bid strategy, you may also want to increase the range (for example, change a range of 6 : 10 to 1.5 : 10).
  • You can also watch the trends of these percentages and adjust bid settings accordingly.

Current keywords

The number of keywords currently in a bid strategy (not the number of keywords during the selected date range). You can click the column header to sort by number of keywords, which allows you to focus on the performance of your most active bid strategies.

Current product targets

The number of product targets currently in a bid strategy (not the number of product targets during the selected date range). You can click the column header to sort by number of product targets, which allows you to focus on the performance of your most active bid strategies.

Diff from target pos

A bid strategy's performance against a keyword position target, whether that target is part of a CPA goal or a standalone goal.

  • Negative values indicate that you're exceeding the upper boundary of a position target.
  • Positive values indicate falling below the lower bound of a position target.

This is not simply Average position - Target position. Instead, we compute the error values for each keyword, and then compute the average (weighted by impressions).

For example, consider a bid strategy with a target position range of 1.0 - 4.0 for one keyword. The bid strategy runs for two days. If the keyword receives 1,000 impressions with an average position of 2.0 on the first day, and 500 impressions with an average position of 5.5 on the second day, the average position for the bid strategy is:

  • (2.0 x 1,000 + 5.5 x 500) / 1,500 = 3.83

3.83 is within the target position range, so you might think the error is 0. However, in reality one-third of the impressions were not in the position range.

The new column adds up keyword-level error values, weighted by the keyword's impressions:

  • (0 x 1000 + 1.5 x 500) / 1,500 = 0.5

We entered 0 for the first day, as the position of 2.0 is within the target range of 1.0 - 4.0. We then took the difference from the second day position (5.5) to the lower end of the target range (4.0), and the result is 1.5. We then weighted it by the number of impressions (500) and divided the whole term by all impressions (1500).

0.5 as a number may not be meaningful. However, it's useful when you sort your position strategies.

  • If you see a non-zero number, then you have keywords in the bid strategy that fall outside your target during the selected time range.
  • If the number is higher, then the number of impacted impressions or keywords is higher during the selected time range.

Optimal cost

Based on the bid strategy settings, how much you should have spent for conversions on the biddable items. For example:

  • For CPA strategies, optimal cost is the target CPA × number of conversions (for example, a CPA target of $10 × 100 conversions = $1,000 optimal cost).
  • For ERS strategies, optimal cost is the the target ERS × amount of revenue (for example, an ERS target of 15% × $10,000 revenue = $1,500 optimal cost).
  • For Floodlight strategies, the CPA and ERS targets are evaluated on an activity-by-activity basis. For example, you're managing a set of keywords related to car rentals, and you set up Floodlight tag bidding for car rentals, hotel rooms, airfare, and cruise packages. DS calculates the optimal cost as follows:
    • 20 car rental conversions × $9 CPA goal = $180 optimal cost
    • 5 hotel room conversions × $20 CPA goal = $100 optimal cost
    • 5 airfare conversions × $30 CPA goal = $150 optimal cost
    • 1 cruise conversion × $700 CPA goal = $700 optimal cost
    • The total optimal cost is $1130

Optimal cost is calculated daily based on that day's metrics. For example, if you change your target today, DS doesn't update the optimal cost from last week. Optimal cost always reflects the status of your settings on a given day.

DoubleClick Search displays an Optimal cost of 0% for bid strategies with Position goals because there is no associated CPA or ERS.