Mike Rhodes | AdWords Audit Cheat Sheet

This document is an Adwords Audit Cheat Sheet that can be used to effectively audit Adwords accounts by someone other than the person running it.

===
System Architect: Mike Rhodes
Website: www.websavvy.com.au
Generated as part of the www.BusinessSystemsSummit.com

System Details

Action: Every account should be audited at least quarterly, by someone other than the person running it.

Goal: List insights, recommended actions and expected business impact.
 

Step 1: Start by looking at a high-level overview of the account

  • Date range
    • Think about what date range should be used for the audit
       
    • Look for consistent data over time. Don’t include periods that contain vastly different types of data. Find consistent recent data covering six months to a year.
       
  • Conversions

    • What’s being tracked and what is the measure of success?
       
    • Are macro and micro conversions being used? Macro examples include a lead-gen form being filled out or the sale of a product. Micro examples include a site visitor who looked at more than five pages.
       
    • Is this AdWords account using ‘conv’ and ‘all conv’ well?
       
    • Are values being used well?
       
    • Which metric is used to judge success?
       
    • Which attribution models are being used?
       
  • Goals

    • What does winning look like? What is the measure of success?
       
    • How does this relate to the business itself? What the business goals?
       
    • What’s the revenue model of the business? These can depend on the sales cycle and the type of business.
       
    • What’s the breakeven amount? Is there enough wiggle room to improve profit by allowing the cost per sale to be higher?
       
  • Structure

    • What campaign types are/are not being used? (search, display, video, shopping, DSA, call only)
       
    • Does the naming convention make sense?
       
    • Why are multiple campaigns being used?
       
    • Which search networks are being used?
       
    • What’s the 80/20 of campaigns? Which are the top 1 or 2 campaigns to focus on?
       
  • Settings

    • What campaign settings are being used?
       
    • Are there bidding strategies being used?
       
    • which ad delivery options are being used? Standard or accelerated?
       
    • Which ad rotation options are being used? Are ads being rotated and split tested properly?
       
    • Is there tracking being used?
       
  • Automation

    • Are AdWords scripts being used?
       
    • Are auto-rules being used?
       
    • Are external tools like Optmyzr or PPC Samurai being used to help make decisions and make changes at scale?
       
  • Labels

    • Are labels being used inside the account? Are they being used well? Is a sensible naming structure in place?
       
  • Account level

    • Is the business using any kind of data feeds? Examples include dynamic remarketing and customised ads.
       
    • Do they have negative keyword lists being applied to individual ad groups or campaigns?
       
    • Are placement exclusion lists being used?
       
    • Is account-wide URL tracking being used?

 

Step 2: Segment the data – dig a little deeper

  • Impression Share
    • What’s the overall feel of the potential of the account?
       
    • Is it constrained by budget or ad rank?
       
    • Are there quick wins to be had? If something is being constrained by budget and it’s profitable then remove the budget constraint if possible.
       
  • Top v Other
    • Where do the ads show – top of the page or other (bottom)? Is there potential to increase average position?
       
    • Segment the data by a device using Excel or a similar app. What does top v another look like by device?
       
    • Chart the ratio of top v other to understand what’s trending and what’s changed over time.
       
  • Devices
    • Which devices are ads being shown on? Which campaigns are running on which device?
       
    • What’s the performance difference between the different devices?
       
    • Are bid adjustments being used?
       
    • Are there quick wins by bidding devices up or down, even at just a campaign level?
       
  • Networks
    • Are search campaigns separate from display campaigns?
       
    • Is the Search Partner Network being used? If so, on which campaigns?
       
    • Are there any quick wins to be had? Examples include turning partners on or off.
       
  • Geography
    • Which locations are performing particularly well, or are performing badly?
       
    • Are they being bid differently? Is Bid adj. being used?
       
    • Are any locations excluded?
       
    • Check if there’s a big difference between where the user is vs what they’re searching for.
       
    • Are there quick wins to be had by changing bids?
       
  • Time
    • Look for seasonal patterns across multiple years or across the last 12 months.
       
    • Are there patterns in ‘day of week’ or ‘hour of day’? Does traffic get very expensive at certain times of the day?
       
    • Are these insights being used to bid differently?
       
    • Do some days/hours need to be bid up or down?
       
  • Experiments
    • Are regular campaign experiments being used and being used well?

 

Step 3: Get granular – delve into the nitty-gritty of AdWords

  • Brand
    • Are there brand campaigns set up? Are brand keywords being bid on? Are they being bid on properly?
       
    • Are those brand keywords being bid on in separate campaigns?
       
    • Do brand campaigns have great impression share and very high positions?
       
  • Ad Groups
    • What structure is being used? For example, Single Keyword Ad Groups (SKAGs) - one keyword per ad group.
       
    • Follow the money. What are the top ad groups by cost? Where are most of the money being spent?
       
    • What's the range of keywords in those top groups?
       
    • Are there big differences in Cost Per Action or Conversion Rate or ROAS?
       
    • Use a pivot table in Excel to work out how many ad groups there are per campaign. This will give you a high-level overview.
       
  • Keywords
    • How many keywords are there in general?
       
    • How are they distributed throughout the account? How many keywords are there per ad group?
       
    • Which match types are being used? Does this account have lots of broad keywords, or mainly exact keywords, or a mixture?
       
    • What’s the relative performance of those different match types?
       
    • What are the top 1-5% of keywords? What are they and how are they performing?
       
    • Are there big differences in the main measures of success – CPA & CR%?
       
  • Negative Keywords
    • Are negative keywords being used well?
       
    • Are negative keyword lists being used?
       
    • How often are negative keywords being added?
       
    • Are there easy wins to be had by adding more negative keywords?
       
  • Search Queries
    • How closely do search queries match the keywords? Are ads being shown for something completely different than the account owner was expecting? If so, that could be a problem.
       
    • Are there search queries that could be added as positive keywords as well as negative keywords?
       
    • Are new search queries being discovered over time?
       
  • Quality Score
    • How many keywords are there with impressions for each quality score? Chart the quality scores for all keywords with impressions to understand the overall quality of the account.
       
    • Roll that up into one big number to see the quality score weighted by impressions.
       
    • What’s the history of the quality score?
       
    • What components of the quality score are consistently below average? This gives you a clue about what needs to be worked on the most to improve the quality score.
       
  • Ad Copy
    • Are ads being tested, or are they being split tested?
       
    • Are big differences being tested?
       
    • Overall, how many ads are there per group?
       
    • What type of ads is being used? Legacy ads should not be used unless their performance is really good.
       
    • Are headlines (1 & 2) being tested?
       
    • Are paths being tested?
       
  • Extensions
    • Check that all or some of these ad extensions are being used:
      • Expanded site links
      • Callouts
      • Structured Snippets
      • Price extensions
      • Message extensions
      • Review extensions
      • Location extensions
      • Call extensions
      • Auto extensions
    • Every account should at least have expanded, callouts, structured snippets and price extensions (where applicable).
       
  • Final URLs
    • Which pages on the website are the ads pointing to? Is the homepage being used too often?
       
    • Is there a wide range of landing pages?
       
    • Does that list look sensible? Are there big differences in the landing pages, or could some be combined?
       
    • Is UTM tracking being used and are they being added to the URLs in a consistent manner?
       
    • Visit the landing pages and audit those too (if time allows).
       
  • Audiences
    • Are remarketing lists (RSLA) being used?
       
    • Are they used at campaign level or an ad group level?
       
    • Are demographic bids for search ads being used? If not, could they be?
       
  • Shopping
    • Are shopping campaigns being used?
       
    • What’s the ROAS/CPA relative to the other campaigns?
       
    • What shopping platform is the site built on?
       
    • How many different shopping campaigns are being used?
       
    • Are the campaign priorities being well used?
       
  • Dynamic Search Ads
    • Audit the DSA campaigns. Are they being used?
       
    • What do auto targets look like?
       
  • Call Only
    • Audit the Call Only campaigns. Are they being used?
       
    • Are they working well?
       
    • Are they set up well?
       
  • Display Network
    • Which targeting methods are being used?
       
    • Can you map campaigns? Use the free tool at TheDisplayGrid.com to help you think about ways that you can target your display ads.
       
    • Are there placements that could be excluded?
       
    • Are ads running on mobile apps?
       
    • Are bid adjustments being used?
       
    • Are placement exclusions or exclusion lists being used?
       
    • Are category exclusions being used well?
       
    • Is there a lot of testing being done between text ads and image ads?
       
    • Are responsive ads being tested?
       
    • Are Frequency Caps applied to the various campaigns?
       
    • Are Smart Display Campaigns being used?
       
  • Remarketing 
    • Are multiple RM lists being created?
       
    • Are multiple durations being used?
       
  • YouTube
    • Are YouTube campaigns being used? If not, it can be a missed opportunity.
       
    • What’s the performance like relative to others?
       
    • What’s the goal of these campaigns?
       
    • If YouTube is being used, are remarketing lists being built?
       
  • Analytics (this is not a full GA audit)
    • Is there a view for raw data?
       
    • Are filters being used?
       
    • If it’s a retail site, is enhanced ecomm being used?
       
    • Is the currency set correctly?
       
    • Is the right time zone being used on Analytics and AdWords?
       
    • Is site search turned on?
       
    • Are display features enabled?
       
    • Are Analytics and AdWords linked so that data flows properly between the two?
       
    • Are goals well set up (if micro conversions need to be tracked)?
       
    • Are all appropriate goals being imported into AdWords properly?
       
    • Are goal values used well?
       
    • Are events bring tracked well? Are those events being turned into goals?
       
    • Are Analytics remarketing audiences being created?
       
    • If so, are these Analytics remarketing audiences being imported and used in AdWords?

Videos

Tags