All forms of paid advertising use the max bid – it’s the most basic concept yet calculating the theoretical maximum value to pay for a click is surprisingly difficult. It’s not that the math is tough; it just takes stepping back from the granularity of channel specific optimization to get a more holistic look at how marketing channels interact with each other.

Looking at one marketing channel at a time provides incomplete insight into performance yet, ironically, many businesses split management of channels across multiple teams and incentivize marketing mangers on performance leading to internal competition for the same conversions.

For example, let’s consider a Google paid search account over the course of Q4 2011. AdWords gives us the basics key performance indicators:

Because AdWords (or any other channel using a proprietary cookie/pixel based tracking system) takes credit for all conversions that were touched by a Google paid search click at some point in the conversion funnel, conversion values reported in AdWords always skew high.

If we compare the same date set within our analytics platform, you will see dramatically different results:

In this case, the conversion count in analytics is significantly lower (about 33%) – enough to make the channel appear unprofitable. This is due to a last entry attribution model that is used by default in most analytics platforms.

So which one is right? It depends on how we want to interpret the data…

What distinguishes an analytics platform from is that it monitors multiple referring channels whereas a channel specific platform (eg AdWords) operates in it’s own little world without any insight or concern for other marketing channels with respect to conversion counting.

This often leads to double counting of conversions if we were to add up conversion counts across all paid traffic channels (AdWords, AdCenter, Facebook ads, etc.).

As a result, most analytics system offer some form of a multi-channel attribution system or conversion funnel. Whether it’s a slick visual graph like in Google Analytics (below) or a simple linear attribution report as provided by Omniture, the idea is to identify the true value of a click or a visit by parsing the value of each conversion across all touch points leading up to the conversion.

 

In order to identify the true value of a click from a given marketing channel, we need to gather some channel specific data:

  1. Clicks
  2. Conversions
  3. Average Order Value
  4. Number of Channels per conversion

Note: I prefer to use an average order rather than revenue in order to normalize the potential effect of a large order skewing data.

Clicks and average order value can come directly from analytics or the channel tool. Conversion count comes from isolating the number of conversions attributed to the channel through the dividing the number of conversions by the number of unique channels associated with each conversion.

To get these numbers, we need to use the multi-channel conversion funnel report and a few different attribution models:

  1. First Click
  2. Last Click
  3. Linear

By multiplying AOV and conversion count (for the two high and low attribution models with respect to conversion count) and then dividing by the number of clicks we can get a range for revenue per click.

Consider the following conversion count per attribution model:

  • Last Click: 2,975
  • First Click: 2,399
  • Linear: 2,203

The high is last click (2,975 attributed conversions) and the low is Linear (2,203 attributed conversions).  Running the calculations:

(2,975 conversions x $25) / 27,348 clicks = $2.01 / click

(2,203 conversions x $25) / 27,348 clicks = $2.72 / click

This tells us that we can afford an average CPC between $2.01 – $2.72 for Google paid search to run at break-even profitability.

The value of a click is completely dependent on the average order value. If your business has significant seasonal fluctuation in terms of search volume, competition, and/or average order values, I strongly recommend running through this exercise multiple times per year in order to best understand the current max bids to operate a profitable marketing campaign.

Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.

Related Topics: Channel: Analytics | Search & Analytics

Sponsored


About The Author: is in charge of client strategy at Fusion Tree, a performance marketing and analytics consulting group, and is based in the San Francisco bay area.

Connect with the author via: Email | Twitter | LinkedIn



SearchCap:

Get all the top search stories emailed daily!  

Share

Other ways to share:

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • http://www.rimmkaufman.com George Michie

    Hi Benny, this strikes me as an okay approach for a small program, but far too crude for someone spending real money in search. If the audience is folks spending < $5K/month, I'm with you, the data is likely too thin to do much else with it.

    However, for the folks we work with (multichannel retailers, travel, financial services, other) spending $50K to $2 or $3 million a month your approach makes much less sense. The value of traffic does not depend strictly on AOV, but also on conversion rates. "Cheap flight to NYC" will have a very different conversion rate, a hugely different AOV and likely a different type of cross channel interaction profile from "New Zealand Vacation Packages". Granularity matters, and parsing credit between channels and within channels at the order level as we do is essential to growing a program profitably.

    Again, what is good advice for one type of program is often nonsense for another type so don't mean to throw stones, just making the case that in large programs with richer data sets a more granular approach will produce much better results than looking across the whole program with averages.

  • http://www.esearchvision.com Benny Blum

    @George – thanks for the comment. You are correct that this is a relatively crude analysis and that there are ways to dig out more accurate values per click for different types of KWs. However, I think your jumping ahead and making some assumptions about the validity of a basic analysis versus a more in depth look comparing upper funnel lower funnel keywords or informational versus transactional searches. What I’m saying is that you’ve got to start somewhere to understand if you’re under or over valuing given channel before spending a lot of time to parse out your KW set by conv rate or user intent. From there, you can parse out sub-sets of KWs to identify if the over/under valuation varies by intent or KW type.

  • http://www.rimmkaufman.com George Michie

    Thanks for your response, Benny. You’re right, you have to walk before you can run, and you also have to work within the constraints of the tools available to you. This is a good method of getting a sense of how the program is really doing overall. That said, ultimately you need these insights to inform action, and the mechanism for taking action is through bidding which is done at the granular level. At the end of the day if the more accurate value measurement doesn’t feed back into the bid management system what’s the point?

  • http://www.esearchvision.com Benny Blum

    George – the validity is rooted in effectively manage up within your clients organization. Complicating the value of one click versus another can easily go over the heads of non marketing- savvy individuals but a simple dollar amount can go along ways to justifying dollars spent on the marketing side. To that effect, an interesting analysis would be aggregating by user intent (informational v navigational v transactional) even though it, as you suggest, has no bearing on bid management. Instead it provides insight into an expected and justified cost per click for a given channel.

  • http://www.rimmkaufman.com George Michie

    Good point, Benny, thanks!

  • http://www.investmentpropertycalculator.com.au/ patrickshi

    (2,975 conversions x $25) / 27,348 clicks = $2.01 / click

    (2,203 conversions x $25) / 27,348 clicks = $2.72 / click

    Are these calculations correct?

    Furthermore, I think this should be based on profit not revenue.

  • http://www.esearchvision.com Benny Blum

    @ patrickshi – thx for the comment. Seems that the answers to the equations were just flipped. The range is still the same.

    Regarding profit vs revenue: good point. Same process applies. Just multiply the click value by your margin. Be sure to normalize the margin to take into account this is a top line analysis. As George has previously suggested, when you dig in deeper to parse out different types of KWs or which types of terms drive sales of different products, you’ll be able to more accurately show the value of a click for a smaller sub-set of KWs.

  • gbfhsghtr

    You can find many Burberry Outlet Store Online now with the google website search ,so that you can buy some Cheap Burberry Bags directly from the website,you needn’t go aboard now if burberry store are not exist in your local city.And 
    the Burberry Bags On Sale with a incredible price ,you should be happy with them.

  • http://profile.yahoo.com/N4XXEX4H35MTIB77WT3I3LIEEA Irina

    HI Benny,

    Where do I take these First Click, Last Click, Linear from? I’ve searched everything I could in GA, and it seems like First Click = first click conversions, last click= last click conversion, but where is Linear? Thank you,

 

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!

 


 

Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide