ClickEquations Blog

Revenue attribution is how you (or the software tracking your online marketing activities) decides where to place the credit for the sales that occur on your website.

If someone who has never been to your website before does a paid search, clicks an ad caused by a keyword you bought, then makes a purchase, the attribution is easy. They keyword that you paid for to get the click, should get credit for the revenue generated by the purchase.

But very often, this is not the scenario that leads to conversions which take place on your site.

• People come multiple times before purchasing.
• They often come from different sources each time the come, occasionally repeating sources along the way.
• They sometimes make a series of purchases, either after all of their visits or interlaced among their visits.

These are just a small sampling of the issues and don’t begin to define or describe the complexities.

And this is not the post where I’ll try to do either. (Those will come.)

But was we work to sort out the right way to handle revenue attribution within ClickEquations, we’re capturing some data that I thought it would be interesting to share.

The following images document real-life ‘click-chains’ – sequences of visits to a website with resulting or intersperced conversions. They are a tiny tiny fraction of the sequences found in one account in a 30 day period.

• Rows with green ‘P’ cells are visits that came from paid search keyword clicks.
• Rows with white ‘N’ cells are visits from organic search, email, affiliate links or other sources.
• Rows with yellow ‘C’ cells are conversion events.
• The number in the first column represents the visit number for that person over all time.
• Only visits within a 30 day window are included although the visit count may have begun far earlier.

And if you’re into this kind of thing, they’re very interesting.

Click-Chain Histories

#1 – Our first contestant is a frequent visitor (note we’re starting with visit 37), loves those paid search ads, but does buy at least occasionally.

Four more after the jump (as they say)

Continue Reading |Comment

The issue of how to allocate credit across the different searches (or other visit types) that lead up to a conversion event is a deservedly hot topic.

This is true largely because the broadly used ‘last click’ allocation model (where the last keyword gets 100% of the revenue credit) is really inaccurate and inadequate.

There are several other attribution models, including first click, linear, weighted, and other hybrids. The pros/cons of each are worthy of extensive discussion. That’s not the topic here however. We have a related question and would like to get some feedback.

Should the allocation of credit ever extend beyond a single conversion?

Two examples below further frame the question. In both cases, assume there is a user-defined tracking period which applies. In other words if you’ve set a ’30-day tracking period’ in any case the look back for events would only extend back 30 days.

CASE 1

Search on KW1 –> Search on WK2 –> Conversion 1 –> Search on KW3 –> Conversion 2

Question: Should KW1 and KW2 be part of the allocation chain for Conversion 2? If so KW1 and KW2 could get partial revenue credit for both Conversion 1 and Conversion 2.

Or does Conversion 2 only allocate back to KW3?

CASE 2

Search on KW1 –> Search on KW2 –> Conversion 1 –> Conversion 2

Question: Should KW1 and KW2 be part of the allocation chain for Conversion 2?

Or should Conversion 2 be seen as an independent event (perhaps inspired by a follow on email or other interaction after Conversion 1.)

FYI, Google Adwords does allocate revenue to keywords from multiple conversions if no search happens between them.

Your Opinion?

We’d love to hear your opinion on this. Please take the survey below, and leave any other thoughts in the comments.

[wp_surveys]

Seen some crazy broad matches lately?

Everyone has, and PPCProz is building a list of the zaniest, so we can all laugh our way to the poor house.

There will always be broad match in every campaign. But if huge portions of your traffic/revenue are coming through broad match you (or your agency) are not working hard enough.

And if you have a lot of broad match it’s certain that you’re:

• Paying too much per click
• Missing a lot of impressions
• Getting useless clicks
• Wasting money

I’ve written extensively about Match Types in the past, and proposed the ‘Match Type Keyword Trap‘ as a guiding principle of how and when to use them and most importantly transition your search queries into phrase and exact keywords, with appropriate bidding.

Broad match is an important tool – it saves time and energy, and provides a place to start.

But mostly – although not entirely – it is a training wheels set you should get beyond for the lion’s share of your PPC spend and revenue.

Average Position occupies an important place in the mythology of paid search.

Many people covet or chase higher positions, and there are several possible reasons:

1. The assumption that ads in higher positions get more clicks simply because they’re in higher positions.
2. As we all know ‘higher is always better’ – especially when it costs more.
3. And of course, eye tracking studies prove, um, er, that people look higher more often.

(The empirical and anecdotal evidence I’ve seen suggests that the power of higher positions is much less than most people seem to imagine. In a future post I’ll go into this in great detail. This is not the real subject of this post.)

As a result, there is a lot of attention paid to the Average Position metric. And a LOT of money is spent on upward bid changes made because of the number this metric reports.

So how good is this number? Probably not very good.

Let’s take one tiny little case study to demonstrate.

The keyword is ‘cat treatment’. And on Saturday Jan 3rd it produced about 25 clicks in one of our accounts. The average position for the term (in broad match) was listed as 4.55. This is the average of all the positions in which it appeared during the 1543 impressions it enjoyed that day.

Now be honest, despite all you know about averages (including the fact that it could have appeared in position #1 760 times, and in position 8 783 times) when you see that 4.55 was the average it makes you think it spent the day bouncing between position 4 and position 5. Right?

But did it?

Let’s look at Google Analytics’ handy Keyword Position report for this keyword on that day. This shows the position the keyword was in when it earned its 25 clicks.

Yowza!  This keyword covered more ground than Paris Hilton in NYC on Saturday night. (I always wanted to see how much Google traffic a single Paris Hilton reference caused.)

It was in all three top positions, and everywhere on the right side from position 1 to position 6.

Keep in mind that this is a map of clicks, not impressions. So maybe the impressions did cluster closely around the 4.55 average and the few stray impressions way up to top 1 and down to position 6 just all got clicks. Or maybe the actual impression distribution was extremely broad and the 4.55 average, while it is true, is really not useful to us in terms of analyzing keyword performance or making bidding decisions.

At this point only two things are really clear;

1. We really need better information. If the search engines won’t provide the actual impression and click position distributions, and/or make the the position-at-time-of-click a macro that can be delivered in the target URL, they should at least provide the standard deviation for the average position so we have some idea of what it really means.
2. We should resist the urge to put much faith, or make too serious of decisions, based on the reported Average Position of any keyword.

Here’s a wild one to end the year:

One of our clients bought a keyword, got a click, and made a sale. So far so good.

Actually they made three sales, to the same person who clicked on that one keyword.

• One conversion was during the visit following the click.
• One was made an hour later when they returned and bought some more.
• And then again 7 days later and bought even more.

Here’s the surprising part: Google Adwords took revenue credit for all three sales as one conversion and applied the total as revenue on the day of the click.

The second conversion could have come from click on a follow-on offer in the first purchase confirmation email. The third one could have come from a click on a banner ad the person saw 6 days later. Should that first keyword get the revenue credit for all three sales?

There is no easy answer.

But this does end the year where I think we’ll spend a lot of time next year – improving both the understanding and practice of revenue allocation.

All paid search campaigns are an effort to gain some type of return; we spend in hopes that we get back more, either in terms of gross revenue or net profit. If we can’t measure how much we get back, and/or if we can’t easily and accurately associate that revenue with the keywords or other sources of traffic to the site, we can’t measure our return correctly. And if we can’t measure return correctly we can’t come to conclusions about our current efforts or make decisions about what to do next.

It’s a giant problem and the stakes in PPC have risen to the point that we can’t ignore it anymore. I know this blog will devote a lot of attention to it in 2009 and I believe it will become a common theme in the industry.

After all the recent attention on Quality Score, I had planned to turn my attention to bidding, and write a series of posts on this important PPC topic.

But I think Ad-Rank deserves a little attention first.

Ad-Rank doesn’t get very much attention – certainly a lot less than bidding, and even a lot less than Quality Score. But Ad-Rank determines your position, and to some degree whether your ads display at all.

Ad-Rank Defined
According to Google, “Ads are positioned on search and content pages based on their Ad Rank. The ad with the highest Ad Rank appears in the first position, and so on down the page.”

Ad-Rank = CPC bid (Max CPC) × Quality Score

So we bid to gain Ad-Rank.

And we care about Quality Score because (among other things) it helps us achieve Ad-Rank.

Ad-Rank and Quality Score
Quality Score is important because it is weighted equally with your bid in determining where/if your ads run. The two factors are intertwined and the result is interdependent.

As the chart at right shows, you can get the same Ad-Rank with a lot lower bid by improving your quality score.

And as Quality Score gets more important, bidding gets less important. Not unimportant, but less important. It’s a zero-sum game.

Ad-Rank and Bidding
Yet while Quality Score is getting more visibility and mind-share than ever before, I’m not sure its ascent is being considered when thinking and acting on bidding.

Creating bidding strategies and running bidding rules or using bidding algorithms that don’t take QS into account at all, seems strange and seriously sub-optimal.

Traditionally bids are decided in an effort to impact position, and often the assumption is made that when an increased bid resulted in a higher ROAS or ROI, it was the change to the bid that was the direct cause – but I don’t know of a rule or algorithm in any PPC software today that either a) checks the actual position impact or b) checks to see if a Quality Score change was a mitigating factor.

It should of course now be noted that both bid and quality score increases have other impacts beyond their influence on Ad-Rank; Google has said that minimum bid and quality score thresholds are set for achieving Top (as opposed to Right Column) positioning, for example. So there are cases where it’s wise to increase your bid regardless of Quality Score issues.

Ad-Rank and You
The core tenant of High Resolution PPC is that we’ve all been lulled into an over-simplified view of how paid search works, both to accelerate our adoption, simplify our understanding, and keep us from complaining about really unfair or opaque aspects of the system that is taking our money.

Ad-Rank is an open secret. It’s well documented, easy to understand, extremely important, and almost never discussed. Time to change that.

A lot of advertisers have keywords on which their desire and instinct is to ‘bid to position’ – meaning they want to rank in the top slot (or the top 3 slots) and are willing to pay almost anything to do so.

This is generally defined as a ‘branding’ requirement, although it may be more accurately described as a form of vanity bidding.

In preparing for next Tuesday’s Quality Score seminar with Bryan Eisenberg, I’ve started thinking about the impact of Quality Score on Bid-to-Position.

It’s easy to think about Bid-to-Position is based on the out-dated thinking of PPC as a pure auction. The strategy itself implies a willingness to pay ‘whatever it takes’ to attain a certain position in the rankings.

(Or at least pay an amount more than economically justifiable – many Bid-to-Position rules do allow you to set a MaxCPC over which the desire for a certain position will yield to some economic reality.)

But as the role and impact of Quality Score increases the ability to bid your way into a position gets harder and harder, and in many cases ultimately impossible.

Position is not driven solely by bid anymore. And in many cases bid won’t even be the largest influencing factor.

We can see this by looking at Google’s new ‘First Page Minimum Bid’ which, according to Google is “based on the exact match version of the keyword, the ad’s Quality Score, and current advertiser competition on that keyword.”

If that’s what it takes to get on the first page, then this obviously has a lot of implications to anyone hoping to ‘Bid-to-Position’.

1. First it reinforces the notion that all of your keyword work should be striving toward Exact Match, because Exact beats Phrase which beats Broad.
2. Second it says you had better really worry about the Quality Score of the keywords you’re trying to position. A lousy QS will sink your chances of attaining any position, let alone a top one.
3. Bids are only important in the context of these first two.

On Tuesday Bryan and I will dive deeply into Quality Score and how you can and why you need to focus on improving it for the keywords in your campaigns. This ‘secret formula’ has impact on every dollar you spend, and every click you get – or don’t get.

The impact of Quality Score on various bid strategies and campaign goals is just one of the topics we’ll cover. Please join us if you can.

Target, Value, Satisfy, Understand. That’s the mantra of High Resolution PPC.

The idea is to stop thinking about mechanical components like keywords and bids, and instead focus on a logical marketing progression.

We want the tools to support our work process instead of having to build a work process that serves the tools.

The First Step is Targeting

Targeting means showing your ads to the right people. Paid search ads are delivered as answers to questions. People type in a search query and you pay for the privilege of having your ad be one potential answer to that question.

So you must know:

• What questions do you want to answer?
• What answers do you plan on giving to those questions.

Campaigns, Ad-Groups, and Keywords are your targeting tools.

Keep in mind that they’re called ad-groups, not keyword-groups. The goal is to segregate keywords, controlled using the match-type option, so that all the queries attracted by a single ad-group are questions answered by the text-ads in the ad-group.

In other words, you want every searcher to see a a text-ad that is directly relevant to their search. To do that, you must organize your ad-groups around the search queries they attract, not the keywords they contain. Every search query that causes your text ads to be displayed, should be highly relevant to the text ad that is displayed.

Let’s illustrate with an example.

Supposed you knew that all of the following search queries would be coming into your account, and you could hand match them to appropriate text-ads before the results page was delivered to the searcher.

• Discount Dyson Vacuum
• Dyson Vacuum Features
• Dyson Vacuum Coupons
• Compare Dyson Vacuums
• Cheap Dyson Vacuum
• Dyson Extra Cyclone

Wouldn’t you want the 3 price-related queries to get a price focused text ad, and the three feature related queries to get a feature-related text ad? Doesn’t it make sense that this would produce the highest click-through-rates and the highest ROI?

Yes, of course.

This is why you have to think about queries not just keywords, and use ad-groups to target the groups of people you want to talk to.

The Second Step is Valuing

Once we’ve targeted the right people using different ad-groups, we can then look inside the ad-group and take advantage of the fact that we don’t have to place the same value on everyone in that group.

Match-Types, Negative Keywords, and Bids are some the core valuing tools.

Extending our previous example, suppose experience tells us that people who search for ‘Cheap Dyson Vacuum’ just don’t buy from us (we’re not that cheap). That has no value, so we add ‘cheap’ or ‘cheap dyson vacuum’ as a negative. But ‘Dyson Extra Cyclone’ is a very specific feature so people who search on that are far into the buying process, we see that query frequently with a high conversion rate. Make that an exact match and bid it up.

You get the idea. By correctly using these tools, watching our search queries and continually refining our campaigns, we can group queries within an ad-group, value them appropriately, and manage both budgets and returns.

The Third Step is Satisfying

People decide how well our paid search advertising does. They decide how to formulate queries which trigger our ads (or not) and they click (or don’t) and buy (or not).

Text-Ads, Landing Pages, and ultimately your offers, website, and checkout process are your satisfaction tools.

When we’re targeting accurately, and valuing properly, we have the ability to focus on satisfying those who see our ads and visit our site. Trying to do so before we’ve completed these steps means, by definition, that we’ve got too wide a range of people coming to really have a fair shot at measuring the results of any attempts at improvement.

There is little doubt that text-ad writing, let alone testing, is the paid search option that gets the least attention and effort as compared to its importance and potential impact. Rewriting a text ad and doubling performance – in terms of CTR which even if it does not improve conversion rate can proportionally increase revenue – is common. We’ve seen many ad re-writes produce 10x-20x CTR improvements. Try that with a better bid.

But writing is hard. Writing is subjective. Writing takes quite a lot of time. None of these make it less important.

All the same is true-er for landing pages, website experiences, and shopping carts. This all very hard, time consuming, and costly work. But it is ultimately directly responsible for the success or lack thereof of paid search campaigns. Even within whatever limitations exist, it should be considered, managed, and measured.

The Final Step is Understanding

Even in this greatly summarized view of the paid search process, there are a lot of moving parts. Each exists by the hundreds, thousands, or hundreds-of-thousands in typical campaigns. They occur tens-of-thousands of times every day as impression and click counts increment. And we have weeks and months of history for all of this to consider and trend.

Paid search can only be managed effectively if you can learn from this data – look into it and find information.

Website and Search Analytics are your tools for understanding.

This means knowing which metrics are important. And when trends are really trends. And how all the numbers affect each other.

It also means that you need the ability to get at the data that can inform you, and easily produce the reports and dashboards that will do so for both you and your colleages or managers.

The key is continuous improvement. Paid search campaigns are never perfect. And they exist in highly dynamic environments. Only through hard work to understand the campaign and know the best move to make next to improve it can you really drive great results.

Summary

The shift into the T-V-S-U mindset is a big one. It changes the process of managing paid search and the way you think about and use the options and tools the search engines provide. More importantly, it aligns your search and marketing goals, and makes it easier to prioritize your PPC efforts and measure your results along the way.

In future posts we’ll dig into each stage and step of this process in more detail. Have questions before then? I’d love to hear them, or your comments.

This post is part of a series on High Resolution PPC, a framework for understanding and managing paid search advertising.

NOTE: This is part of a post series. It’s available as a single post for easier reading: The Match Type Series.
===============================================
In several earlier posts in this series I’ve discussed the how’s and why’s of buying the same or similar terms at the same time with different Match Type settings.

I outlined in one post the details of creating a Match Type Keyword Trap to filter certain search queries into specific match types. Buying multiple terms and multiple levels – when done correctly – has the ability to give you control over which queries are caught at which price.

Exact-Phrase-Broad

One reason this works is because the engines (generally) execute the match types sequentially.

In other words, if you are bidding on the same keyword, or two keywords that would both match for one particular query, an Exact Match should take precedence over a Phrase Match which should take precedence over a Broad Match.

So even though a particular query is technical a match for both one Broad Match keyword and another Phrase Match keyword, the Phrase Match should always ‘win’ and catch that query.

I should hasten to point out, this will not always be true. If you carefully watch query reports for your keywords you will see queries that were exact matches against a keyword you had set to Exact Match, yet the query lands in a Broad Match group. But in our experience these are rare in the sub 1% range of all queries.

Emphasis the Match Type Setting with Higher Bids

You can and should add punch to this precedence by ALWAYS placing rather substantially higher bids on your Exact Match vs Phrase Match, and Phrase Match vs Broad Match when they’re stacked in targeting the same terms.

And make the differences between the bids significant – it generally won’t help to bid $0.05 more for Exact Match than Broad Match. When bidding it’s easy to look at your Max CPCs (since that’s the option used to set the bid) but since your actual and average CPC is usually just a fraction of the Max you really can’t base your decision on those. Look instead at average CPC’s being reported and then set the Max’s at large enough intervals to create real steps between the different keyword/match type combinations.

By placing a substantially higher bid on the match type differentiated keywords, you’re providing another algorithmic reason for the engine to match exact match queries to your Exact Match keywords. Of course, it should also be true that you want generally higher position and higher impression share for the keywords you’re bidding on Exact Match.

A Simple Match

At the start of this series I mentioned that Match Type was a powerful and often under-utilized option. I hope these five posts so far have covered some of the ways you can get more out of these options. Time for a break from Match Type, however. Watch for a new series starting soon.