Author Archives: sdonatelli

All about conversion tracking II

| by Posted by sdonatelli in Blog
Reply

All about conversion tracking II

How tracking scripts work

In part one of this series on conversion tracking, we reviewed the difference between “conversions,” “conversion events,” and “conversion tracking.” In this post, we’ll look at how conversion tracking works, specifically when using tracking scripts.

When you start to set up a conversion tracking model on your web site, you might feel overwhelmed by all the choices that are available. Different systems offer various choices, but there are basically two approaches to tracking conversions – log files and tracking scripts.

We’ll focus on tracking scripts because that’s how most third-party technologies record events.

What Is a Tracking Script?

A tracking script is simply a piece of code you put on your site to record all conversion events. While there are multiple methods and configuration parameters that can be used, most tracking scripts use browser cookies to track website activities.

Here’s how it works:

The diagram above shows a five-step scenario for how tracking conversions and user activity might work on our own searchforce.com site.

Here’s a possible sequence of events:

1.) Visitor searches for a keyword and clicks on www.searchforce.com
2.) Tracking script code installs a cookie on the user’s web browser with a unique visitor ID
3.) The click is recorded by the tracking system
4.) User fills out a form to download a whitepaper
5.) Second tracking code captures the success of an event (whitepaper download), which is also recorded by the tracking system
6.) If the same user returns to the website and performs another conversion event(signup for newsletter) and if the tracking cookie is still in the browser, then the event will be recorded and it can be linked to the initial click.

So the tracking script is simply a mechanism for placing cookies on a user’s web browser, configured to record specific conversion events as they occur.

However, the method for delivering tracking scripts and the configuration options available can vary widely from one solution to the next.

In the next post in this series, we’ll discuss the options that two publishers offer (Google and Microsoft) – and the options available on the SearchForce platform.

All About Conversion Tracking

| by Posted by sdonatelli in Blog
Reply

All About Conversion Tracking

Part 1: Knowing the language

You may have read articles about “conversion tracking,” but it’s probably one of the most misunderstood aspects of online marketing, one that brings to mind many questions. How does it work? What technologies should you use? Why do different systems count conversions differently?

This series of posts will help you understand conversion tracking, how it works, and what options you have on our platform for measuring conversions.

Sure, this may be a refresher for many of you, but there are plenty of marketing professionals who’ve been thrust into a role of managing campaigns without truly understanding how conversion tracking works. And knowing about these issues is essential to running an effective e PPC campaign.

So let’s run down a list of terms that SEM professionals use when discussing these concepts:

Conversion events:

When you track conversions, you’re recording various events that take place when users visit your site. Those actions are called “conversion events.” Examples could be page views, sign ups, or actual sales. Some of these conversion events are worth tracking, while others are not. The decision to track a specific type of event depends on the nature of your business and your campaign optimization goals.

Conversion tracking:

If a user takes a specific action that you’ve defined as a conversion event, the platform you use should record that event. Some events may be more likely to lead to a sale, so you can assign a specific value to that event. For example, a page view might be assigned one value, while a signup might receive a higher value.

Conversion rate:

Tracking conversions can help you measure the effectiveness of various inbound sources of traffic. Comparing your total traffic against conversion events gives you your conversion rate. For example, for many paid search campaigns, a conversion rate might be calculated like this:

Conversion Event/Total clicks = Conversion rate

This represents the percentage of total clicks that generated conversion events. But whether a specific event is an actual “conversion event” depends upon how you set up your tracking system.

Where it gets confusing: Not all conversions are alike

A “conversion” may represent different actions depending upon how you measure what users do on your site. And that can get confusing because some advertising executives may only consider an actual sale to be a conversion, while others may consider intervening actions, such as filling out a form, downloading a white paper, etc. to be conversion events as well.

So there might be multiple conversion events that lead to one conversion, perhaps three different events – or even more, as shown here:

Conv. Event 1 + Conv. Event 2 + Conv. Event 3 = Conversions

And that means each person’s analysis of the conversion rate can be different. Let’s look at an example of how multiple conversion events might be tracked

• You receive 100 clicks on an airline paid search ad
• 98 visit your pricing page – so you have 98 “page views”
• 25 forms are filled in to get airline prices, which you might define as “leads”
• 3 users actually book airline tickets so you have 3 “sales”

So what’s the conversion rate? That depends on how you measure it. You could view it in a number of ways

• 98% (98 page views / 100 clicks)
• 25% (25 leads / 100 clicks)
• 3% (3 sales / 100 clicks)
• 126% (98 page views + 25 leads + 3 sales / 100 clicks)

You can see how this can get confusing. If you refer to a sum of all different types of conversions, you might not be speaking the same language as someone who is mostly concerned with the total number of actual sales.

Most search management technologies consider conversions to be the sum of all the different types of conversion events that you define. So to build a customized plan to optimize your campaign, you’ll need to make sure you know precisely what these terms mean.

So to review:

• Conversion: An action you want someone to take on your website
• Conversion tracking: The practice of tracking and assigning value to the conversion event
• Conversion events: Specific types of events you track and assign values to

In the next post in this series, we’ll discuss some of the common ways people use technology to implement conversion tracking solutions to track specific types of events.

Emerging social influence inside AdWords

| by Posted by sdonatelli in Blog
Reply

Emerging social influence inside AdWords

In the season of sharing, we wanted to pass along an excellent post that our good friend David Rodnitzky wrote on the emerging social influence inside AdWords. We suggest you check it out and wish you all a Happy Holiday!

How Google AdWords is Becoming More Social
Written by David Rodnitzky, PPC Associates
Web source: www.jeffbullas.com

Keyword Quality Scores: Avoiding The Optimization “Bridge to Nowhere”

| by Posted by sdonatelli in Blog
Reply

Keyword Quality Scores:

Avoiding The Optimization “Bridge to Nowhere”

The term “quality score” is probably the most polarizing one in the realm of pay-per-click advertising.

Some practitioners are fixated on boosting quality scores to lower CPCs and raise CTRs. But taking this path can often seem as if it’s an optimization “bridge to nowhere.” Given that quality scores are based on a “black box” algorithm, raising those scores isn’t easy

Other practitioners recognize that black box for what it is and take a contrary view. Why, the thinking goes, optimize against an unknown formula? Quality scores may be good directional data points, but aren’t worth the time and resources for optimization initiatives.

I can understand that second point of view, but overall, I don’t think you can ignore quality scores. There is, of course, no guarantee that optimization will work l, but that black box is actually more of a grey box. We do have some insights into how Google calculates quality scores – and can use these insights to improve campaign performance.

Reviewing quality scores: How Google auctions keywords

This example will explain how I approach quality scores. This distribution shows total clicks and distinct count of keywords from a set of anonymous SearchForce data.

 

As you can see, most of the keywords had a quality score of 3 or 4 and most of the clicks came from keywords with those scores. Does this provide any useful information? Yes, because we know that CTR itself is one of the most prominent attributes for determining this score.

Yes, there are plenty of other factors that impact a specific search query (landing page quality, account history, thematic ad groups, etc.), but Google tells us that CTR is among the most important factors.

To see how this is important, let’s look at how the Google auction actually works. (Wired has an excellent article on the topic of “Googlenomics” if you’d like to dig into this topic in more detail).

If quality scores directly impact the price you pay to clear an auction for a keyword – and CTR represents a leading indicator of relevance – then we should see some relationship between CTRs, quality scores, and the average price paid per click.

So here’s another look at that same data from the chart above, this time looking at average CPC and CTR by quality score.

 

Here we see a strong relationship. As quality scores improve, CPCs decline and CTRs increase.

If you’re cheap like me, this matters. Most advertisers prefer to clear an auction at the lowest CPC possible, without sacrificing volume I realize there are plenty of other metrics that impact campaign success but this chart shows it’s pretty clear that it’s possible to pay less and get more clicks.

Improving quality scores isn’t something you can accomplish overnight, but ignoring quality scores entirely? That’s truly an optimization “bridge to nowhere.”

 

 

Dynamic Search Ads – What Are They? When To Use Them?

| by Posted by sdonatelli in Blog
Reply
Google Adwords - Dynamic Search

Dynamic Search Ads

What Are They? When To Use Them?

Last week Google officially announced the beta of Dynamic Search Ads, a feature they have been testing with select advertisers for a little while now. As Google continuously innovates as it pertains to paid search, it’s important to stay on top of and understand the new ad formats that they roll out.

Depending on who you ask, this could either be the greatest innovation Google has offered into AdWords, or it’s just a sloppy way to build a paid search campaign. But here is what it is. Effectively, a dynamic search ad is a keyword-less paid search campaign. Through Dynamic Search Ads, you specify a website, category, specific URLs or page content that you would like to target ads for. Based on your target method for ad creation, Google will then automatically determine what keywords to show your ads on, dynamically create your ads, and determine which landing page to send traffic to. So, you don’t need an SEM manager anymore, right? Not quite.

While very compelling, and I love the automation component, utilizing this feature or the similar AdWords Express option that is available for smaller advertisers should be done so with caution. Dynamic, keyword less ads can serve a purpose, but probably should be reserved as a secondary method for managing paid search campaigns (dynamic ads, per Google will never serve over a keyword targeted ad in the same account). Some hypothetical instances where this feature may be useful would be the following:

Small Advertisers: This is the perfect local or small business solution, right? Small advertisers often have neither the time nor expertise to properly setup and manage a PPC campaign. This however assumes relevant content, structure, and calls to action on the destination website, which we know is often not the case for large advertisers, let alone the small local advertiser who may struggle to even get a website up in the first place.

Inventory Management: Retailers with many SKUs or product listings that change frequently could find this feature useful as a way to maintain product inventory in synch with paid search campaigns. It is inherently faster to update inventory in your paid search ads via a feed or a crawl than to have a search manager manually research inventory levels and pause/un-pause keywords/ads whose inventory levels are either in or out of stock. However, this is a feature advertisers who rely on inventory management often already utilize, either through an in house developed solution, or through a 3rd party platform that supports inventory management.

Keyword Research: Don’t have time to investigate every keyword that may possibly be relevant for your PPC campaign? Then dynamic ads would be a good way to do keyword expansion and discovery. However, you already have the same option to do this in keyword campaigns. It’s called broad match. You should be employing broad match already as a discovery method for finding new keywords to bid on in a more restrictive match type.

As with any new AdWords feature, it’s important to test as results will vary on an individual case basis from one advertiser to another. But do mistake dynamic search ads as an alternative to a well thought out and articulated paid search campaign with thematic keyword groupings, and relevant keywords pertaining to your business. There are a lot of things you can automate with paid search to achieve better results, but I would argue against giving complete control to a publisher to determine what is relevant to your business and what is not. This is not to say you should not try Dynamic Search Ads, rather, it is to say that if you to try them, they should be in addition to, not in place of, a carefully articulated keyword rich PPC campaign.

Share of Search – Know Your Data: Google Search Share at 66% per Hitwise

| by Posted by sdonatelli in Blog
Reply

Share of Search – Know Your Data

Google Search Share at 66% per Hitwise

The latest numbers from Hitwise are out, and show Google gaining in share of search for September of 2011, to 66%. In the Hitwise report, the Bing/Y! marketplace checks in at 28%, down 2% from earlier this year. As a former paid search manager, I closely tracked both Hitwise and ComScore numbers on search share, as a benchmark to compare to the real internal numbers I saw from my paid search campaigns. The truth is, in some instances the share of search numbers you may see in your campaigns will differ by varying degrees when compared to the reported industry numbers. In my former roles, I always paced Google’s share of search to be in the 70-75% range (often varying with seasonality trends), and seeing the latest Hitwise numbers, I reached out to a colleague to understand what he sees. 75% share for Google.

So, does that mean Hitwise is wrong? Does that mean you should allocate 75% of your budget to Google? Not at all. It simply means the representative data in the industry may not reflect your specific vertical or campaigns, for many reasons. The reported numbers should only be used as a directional proxy for market share. So what does that mean for you, the advertiser, when determining how to manage paid search campaigns, and where to spend your marketing dollar?

Knowing your business, its stated acquisition goals, and the corresponding data your paid campaigns reveal to you is critical. The real truth in numbers often lies in your own data, and can present many good questions that you should be asking yourself? Is Google’s share of search higher than stated industry numbers? How does Google’s share proportionally affect my paid search budget? Does Bing convert better than Google? Does one search engine bring me a higher value customer than another? Does share of search effect my budget allocation and forecasting? These are a few of the many questions that only a paid search manager dialed in to his search campaigns can answer. The real truths to these questions may vary by vertical, geography, brand position, and many other additional factors. Knowing your business’ reality is imperative.

So what does your data tell you? Do you see similarly stated numbers for share of search, and what questions are you asking, that only your data can answer? We would love to hear from you.

Knowing your business, its stated acquisition goals, and the corresponding data your paid campaigns reveal to you is critical. The real truth in numbers often lies in your own data, and can present many good questions that you should be asking yourself? Is Google’s share of search higher than stated industry numbers? How does Google’s share proportionally affect my paid search budget? Does Bing convert better than Google? Does one search engine bring me a higher value customer than another? Does share of search effect my budget allocation and forecasting? These are a few of the many questions that only a paid search manager dialed in to his search campaigns can answer. The real truths to these questions may vary by vertical, geography, brand position, and many other additional factors. Knowing your business’ reality is imperative.

So what does your data tell you? Do you see similarly stated numbers for share of search, and what questions are you asking, that only your data can answer? We would love to hear from you.