Calculating Bid Limits

The first step in analyzing any PPC account or account element is to establish the economic parameters applied to manage PPC ad spending. This analysis can be applied to broad product groups, campaigns, or individual aspects of a PPC account.

Our analysis is framed in terms of quantifiable CPC-type metrics, allowing us to weave into the basic account maintenance process a continuous comparison of our Bid Limits and the Market CPC.

Granularity of Analysis

The granularity of analysis for a PPC account is a matter of a professional PPC Manager’s judgment. Although we often begin analyzing an account at the campaign level, this is not generally appropriate for online stores. Even for service industries, it is often necessary to evaluate specific elements of an account, possibly even keywords separately, particularly when their performances differ significantly from the average for the campaign or ad group in which they belong.

In any event, it is always necessary to analyze distinct services and products separately.

Special Treatment for Star Keywords

Aside from distinct services and products, it is sometimes necessary to analyze ad groups or specific keywords separately (meaning to calculate a bid limit separately for an ad group or keyword). When a particular keyword, or group of keywords are performing substantially better than the campaign or ad group in which they belong (usually measured by cost/conversion), then a PPC Manager would likely move the keyword, or group of keywords into their own ad group, or possibly even campaign. In the case of a separate ad group, this would be done in order to more closely tailor ads and landing pages to the star keyword(s). In the case of a separate campaign, this would be done to provide a basis for pouring fuel onto the star keyword(s) by unfettering any budget constraints which might have otherwise contributed to a loss of impression share.

Repair the Dogs or Shoot them

If particular keywords are under-performing, then a professional Ad Manager may analyze the underlying reasons for the poor performance, and if the reasons cannot be remedied, then the Dog keywords or possibly even ad groups are paused.

Velocity of Re-evaluation

A professional PPC manager is applying her judgment and experience continuously to determine the velocity of our analytical cycle as it applies to each level and element of an account. Bid Limits are reviewed and possibly revised at least every reporting period, and the comparison of market CPC and bid limits is considered daily, as bids are revised.

Exceeding Bid Limits

The only rational basis for ever exceeding Bid Limits would be when the underlying assumptions for the bid limits do not apply to a specific element of an account. This could occur for a star keyword, for example. When this occurs, a separate bid limit must be calculated for the star keyword (broken into a separate ad group as described above). In other words, never exceed Bid Limits.

Bid Limits are not Maximum Bids!

Sometimes clients confuse our Bid Limits, with Google Adword’s Max Bid input. These are not the same thing. Interchanging the two will incur an eventual loss, and can incur an immediate loss when several competitors in an ad space adopt such an approach. When several competitors in an ad space set their default bids to their bid limits, any windfall profits associated with PPC advertising will immediately accrue to the ad agency, Google for example! This is the primary point of our “Buffer” analysis! Don’t squander your buffer! Preserve your buffer! Enhance your buffer!

C O O P E R A T E !!

Don’t rile competition in your ad space

For most ad spaces, there still exists today a comfortable buffer between Bid Limits (results of our analysis) and the market CPC (actual bid amount for the keyword phrases). This difference bolsters the “R” in an advertiser’s ROAS. When each participant in an ad space cooperates to preserve this “R,” then PPC returns accrue to those competing in the ad space. While this fact may seem obvious, it’s not unusual today to encounter ad spaces which have been at least temporarily compromised by two or more Ad Space Cowboys who may have failed to mind their arithmetic, or worse, chosen to ignore it.

Margin by which Bid Limits exceed Market CPC?

[Decision Node A: Measured on spectrum, High or Low]

Assessing Viability of PPC Advertising

The Decision Node [A] in our diagram establishes whether our calculated Bid Limits exceed the Market CPC. This is a quick test of whether PPC advertising is going to work well for a prospective advertiser.

A great buffer would be when our bid limit is $1200 and we find that we can achieve our targeted ad position with a bid of just $2.24. While that may seem like a ridiculously high buffer, we actually do have accounts which have similar buffers. A decent buffer would be when our bid limit is $7.75 and our market CPC is just $2.24. As long as a buffer exists (bid limit exceeds Market CPC), then theoretically, any ad spending will generate a positive ROAS.

History May Not Foretell

When a professional PPC Manager meets someone who says, “I tried Adwords, and it didn’t work for our company,” you’ll notice her shaking her head and smiling. Every week we analyze prospective PPC accounts as part of our free quote process. We quite often encounter accounts which have been created or managed by novices or imbeciles, and those most often generate a negative ROAS for the Advertiser. In such cases, a massive overhaul of the account will most often remedy the condition, and our proposals describe the specific steps we will take to improve the accounts.

When we encounter a professionally structured and managed PPC account with no buffer, then we know it’s going to be a tough account to manage. In such cases, our first recommendation may be to significantly reduce ad spending. Sometimes, even with lower spending, it may not be possible to eek out a positive or attractive Return on Ad Spending (ROAS). When all else fails, we might even consider pulling the plug on PPC ad spending altogether. Although we sometimes abandon ad spaces (paused campaign spending and even discontinued advertising entire products), we have so far only once ever encountered an entire ad account for which pay per click spending could not provide a positive return on ad spending (and I believe that account could have been salvaged had the Advertiser improved his landing pages and web offerings according to our recommendations).

Velocity of Analysis

In evaluating whether an attractive buffer exists between bid limits and market cpc’s, it should be understood that, especially initially, the calculation of Bid Limits depends on the estimation of conversion rates. Therefore, it is not always possible to determine the existence of an adverse condition prior to spending the first advertising dollar. For that reason, and also because the Market CPC is set by an auction (and is sensitive to exogenous competitive forces), it’s important to continuously reassess the condition of any ppc account / campaign.

Forecasting Conversion Rates

For a new client, we forecast a conversion rate based on our experience for the industry, conversion types, conversion process, and our Competitor Analysis. [Sometimes we may even be forced to apply our shot in the dark conversion rate par value of 1%.] The point is, our initial forecast is imprecise at best, so requires frequent revisiting during the settling in phase of a new account. Over time an account generates sufficient statistics to move towards a more precise estimate of the conversion rate for each distinct element of an account. Generally by the time we complete the first Monthly Report for a client, we have already improved our initial estimates about conversion rates, which we include in our PPC Manager Roadmap.

Conversion rates are not static, and may be improved or even adversely impacted by changes to a number of endogenous and exogenous factors, as more fully described in “1) Noncompetitive within Ad Space.”

Since many of the factors impacting conversion rates are constantly in flux and under revision, the conversion rate, and the Bid Limit for every primary element of an account require regular reevaluation.

1) Economic Condition = Noncompetitive

“Some other places were not so good but maybe we were not so good when we were in them.”

Ernest Hemingway

“PPC Advertising Doesn’t Work For Me”

Ever hear ex-advertisers whine about how Adwords was a waste of money? Most likely, that company spent a wad of dough, whilst sitting on the Noncompetitive node of our PPC Ad Management Process Diagram. They spent it before reaching the conclusion that enterprise value had been destroyed by buying clicks which didn’t convert to new business. If you are the fearful type, read this section carefully, for herein are described those PPC mistakes you endeavor to avoid.

When an advertiser or new account manager finds himself on this node of our PPC Ad Management Process Diagram, a common initial reaction is “I’m screwed!” Well, there is some relative element of truth in that reaction, but in most cases, it doesn’t have to be a fatal ad space condition. Indeed consider yourself lucky for having discovered the condition before busting the bank on underperforming ads.

Improve Performance

Our experience shows that the competitiveness of all ad accounts can be improved. Some have the potential for massive improvement, others for only marginal improvement. If you ask us to review your ad account, in connection with providing a quote, then as part of that process we are going to offer our candid assessment of the potential for improving your account. If you’re already doing a great job managing the account, then we won’t hesitate to say so. The last thing we wish to do is to raise expectations which we cannot confidently meet. If we can recommend some improvements, we’re going to mention those. If your ad space is quite competitive, then our assessment will mention that.

Some ad spaces are hypercompetitive, and crowded with ads placed by savvy ad managers. Others are hypercompetitive, and crowded with ads placed by buffoons. [FYI, the second ad space is the truly dangerous one!!] Others still remain open for virtual homesteading. The important thing before you get started is to understand the competitiveness of your ad space, potential ROAS, potential for improvement, and then to manage your own account in a manner which makes you richer not poorer. Sometimes this involves making the most of the hand you’ve been dealt, not necessarily the hand you wish you had been dealt. Most often there exists sufficient room for improvement to move you to a happier spot on our PPC Ad Management Process Diagram.

Analyzing Failed PPC Accounts

There are a number of possible reasons why you may not be competitive within your ad space. As previously mentioned some of the factors impacting your competitiveness are endogenous, meaning they relate to things under your control, and some are exogenous, meaning they may not be directly under your control.

Conclusion

If you wish to improve your competitiveness, then focus on improving your endogenous factors. If you improve the first four, then you’ll expect a higher quality score, which will directly impact the financial performance of your account.

Sometimes if you stay on your toes, you might also identify an opportunity for exploiting or influencing exogenous factors!

[B] Spending PPC Budget? Yes or No

Determine if a client is spending the daily budget by reviewing an Adwords report, or by reviewing the online Adwords interface for the past week or so. Often it’s not a simple yes or no answer. If the budget is $75, and the campaign is hitting the budget 5/7 days, then we might answer a qualified “Yes.” Anyway, Google loves informing us when an account could spend more (”Campaign Budget Alert”), so this condition is sometimes  easily determinable from just a glance at the online Adwords interface.

[C] Impression Share? High or Low

Find impression share by running an impression share report. Beware about the sensitivity some types of revisions can have on impression share. When an account is running over 75% impression share it should be carefully trimmed, not hacked! So keep in mind the degree to which impression share might be considered high or low. On the ends of the spectrum, it’s easy: less than 10% = clearly low; over 90% = clearly high.

Make sure the report dates are representative of the account’s current condition. If you’ve made significant changes a week ago, then the impression share report should show activity only for the period since that event.

[C] Impression Share? High or Low

Find impression share by running an impression share report. Beware about the sensitivity some types of revisions can have on impression share. When an account is running over 75% impression share it should be carefully trimmed, not hacked! So keep in mind the degree to which impression share might be considered high or low. On the ends of the spectrum, it’s easy: less than 10% = clearly low; over 90% = clearly high.

Make sure the report dates are representative of the account’s current condition. If you’ve made significant changes a week ago, then the impression share report should show activity only for the period since that event.

PPC Condition 2:   Disengaged

Get in the game!  Don’t sit on poor ad positions, below market bids, and an underspent ad budget.

This PPC condition is coincident with poor ad positions.  It’s possible that for some advertisers, poor ad positions (double and triple positions) as a planned sort of ad position strategy might work.  It might be particularly interesting to consider as an odd ball sort of strategy when a primary ad space is hypercompetitive, or overrun with buffoons.  Most often however, a disengaged condition correlates with a poorly maintained account.

When we first started managing ad accounts four years ago, we could buy clicks in many industries for nickels, dimes, quarters.  Some of those same ad spaces have moved forward to dollars and even fivers.  Some accounts which used to purchase clicks for $5 have since moved north of $25 or $50.    Anybody falling behind on bid maintenance over this period might now inadvertently be purchasing triple digit ad positions.  We generally take exception to positions which fall off the first page, because it can result in the arbitrary display of ads.  This contrasts with being in a bargain hunting mode where we are selectively purchasing those terms and elements which offer the best value.  No doubt, sometimes a disengaged approach can result in the display of ads for only long-tail keywords, but unless the account is structured to exploit this approach, it is unlikely to be trending up on its quality score.

Aside from all that, most often when we find an account which is disengaged, its overall condition is a shambles.  In such cases, it may require a complete restructuring, new ads, and an overall critical reevaluation.  In such cases, historical data might offer only limited value, and we would expect to reforecast conversion rates

PPC Condition 3: Starved for Traffic

When you find yourself on this node in our diagram, think CAUTION! If you’re dealing with a mature account, look back at the history of what has been tried in the past to the extent possible. Why are we not able to efficiently spend the client’s ad budget? As we step out and search for other ad opportunities, we risk spending ad dollars unproductively. Since most bargains are often long tail keywords in core ad groups, stepping into new ad forms most often yields only marginal returns, or worse, destroys value. The velocity of our management cycle revs up to a whir, when we add new advertising account elements. Things we can consider to generate more traffic may include:

• Broaden match types
• Broaden geo scopes
• Broaden ad types
• Improve targeted ad positions
• Turn on Ad Placement network
• Create site targeted campaigns

Many of the above can cause the account to perform worse, not better, so careful monitoring is required. Remedial action is required whenever any of the above elements fail to meet conversion expectations and ROAS hurdle rates.

With search marketing, there exist a finite number of prospects within your geographic market who are typing in your relevant search phrases. You can’t force more people to search for your core search phrases. The best you can ever do is to present your ad to everyone who searches (at your targeted ad position). Considering this, for many advertisers who feel they are starved for traffic, their best approach may actually be to modify their own expectations (decrease their budget!!), and focus on improving conversion rates in the core search element of the ad account.

PPC Condition 4:  Bargain Hunting

As PPC Managers, when we evaluate a new client and our analysis lands the client on this node of the PPC Ad Management Process Diagram, it represents Nirvana!  It is akin to being the first to walk through a vast orchard of Peach trees at harvest time, when they have just become flush with low hanging fruit, and selecting just the most perfect bushel of peaches among the entire crop.  It’s the safest and brightest spot on the Diagram.   From here, if the account is structured efficiently, the only thing required to improve performance is pruning.  That’s what bargain hunting often involves, simply eliminating the worst performing elements (and sometimes displacing them with lower volume, higher return elements).

In the bargain hunting mode, we know there are more searches for our terms than we choose to purchase.  As shown in the following diagram, maximizing number of clicks which we can purchase sometimes involves moving from the high volume / high cost keywords on the left to lower volume, bargain keywords on the right.

The above diagram illustrates the search ad space for Auto Painting and Body Shop Repair in Des Moines, Iowa.  There are a little over 600 searches per day in Des Moines for relevant keywords.  Purchasing the entire market of 600 searches would cost an average of $1,256 per day.  Doug’s Body Shop can afford a budget of just $400 per day.  On the left are the high traffic expensive root keywords like “body shop” which cost $5.00 per click.  On the right are the long-tail keywords, like “nissan auto body repair east des moines” which costs $0.10 per click. The average cost for the entire market is $2.00 per click.

Given his advertising budget, Doug can purchase 83 clicks for root keywords on the left or 320 clicks on the right.  Assuming they are all equally relevant, our experience shows that the long-tail keywords on the right usually convert at a higher rate than the root keywords on the left.  Clearly the long-tail approach will allow Doug to get the most out of his advertising dollars.  If Doug is willing to enhance the granularity of the account by breaking the long-tail keywords into finely grained ad groups, then he can also significantly improve his Quality Score and further reduce his costs.

Some of the tactics which might be considered in the bargain hunting mode are:

• Tighten match types
• Kill content network
• Add negative keywords
• Decrease bids
• Place best performing keywords into separate campaign, with plenty of budget
• Place high traffic suspects in separate campaign, and starve them with budget
• Eliminate all (short too general) high traffic, low return keywords
• Modify the account to target only long tail keywords
• Identify best performing days or times of day to run ads
• Tighten geographic scopes
• Convince the client to increase the daily budget

The severity of the pruning or revision should be determined by how low the impression share is.  Don’t underestimate the effect small changes can have on the account.  An impression share of over 70% involves snipping, not hacking.

Often times, content advertising represents marginal spending, and the effect of turning it off can be to bounce the campaign from Bargain Hunting condition to a Starved for Traffic condition.  A common mistake by novice account managers is to bounce an account back and forth between both conditions in perpetuity.  Of course, the proper solution would be to evaluate the ROAS for content advertising on a stand-alone basis and then recommend either that the budget for the account be increased to accommodate the content ad spending, or that content ad spending be eliminated and in some situations even that the total ad budget be decreased accordingly.

PPC Condition 5:  Ad Spending Equilibrium

Ad budget equilibrium might be the result of a lot of exceptionally good work, or it might be luck. For example, one could achieve Ad Budget Equilibrium by tossing together a shabby account, letting it rip, then setting the budget equal to the average daily spend.

The fact is, this node on our diagram can be a very challenging spot to be in, if you are an ad manager, because whatever changes you implement will unbalance the equation. Yet often changes are required. If the account is super-refined and optimized, then mostly you just have maintenance and regular analysis work to do, but in such a case, the prospects for massive performance improvements may be limited. If the account is in a shambles, then you’ll have to suffer the disruption of refactoring it, and that is sure to land you on a different node on our diagram.