I won’t lie — I never liked mathematics when I was in school. Okay, I really hated math. Yet somehow, I chose to work in what I believe to be the most data-driven form of marketing: pay-per-click (PPC).
According to Google, a whopping 89% of consumers research products online, and 67% of Internet users have gone online to search for a product advertised offline. What does this mean to advertisers and agencies? Digital has become a key component of the consumer ecosystem, and PPC is a big testing (play)ground.
Depending on the nature of your program, PPC campaigns can produce an overwhelming amount of data. Embrace it. Most of the campaigns I have managed over the past eight years have been directly or indirectly driven by return on investment, or ROI metrics. The majority of advertisers I have worked with position PPC as below-the-line (BTL) activities, and allocate their budget accordingly. I would argue that PPC is a through-the-line (TTL) form of marketing. Key performance metrics aside, there is more to PPC than meets the eye.
So how can you leverage the gold mine of data you are collecting beyond the point of conversion? To start, I’m going to walk you through four potential approaches to doing so.
Monitor brand and product interest. Google has developed fantastic tools to help advertisers gauge opportunities and follow the market trends.
Trend Reports Over Time:
Is the consumer interest around your brand, products, and category increasing or decreasing? A Google Trends report can (and hopefully will) lead to a larger discussion about last click attribution and its impact on upper funnel queries — be ready to ask yourself those questions before opening the Pandora box.
Because PPC is data driven, it is very tempting and natural for advertisers to optimize towards conversions and focus on terms that generate a high/positive ROI. I will argue, however, that the majority of advertisers are still attributing the conversion to the last click instead of assigning credit to different touchpoints throughout the search funnel.
This strategy usually translates into a budget centered on brand, products, and longer tail terms. The downfall is that ads are not showing when a user is starting the research process using more generic terms (e.g., “laptop,” leading to “small laptop,” and eventually converting on “laptop 14” brand X”). The questions to ask are:
- How many conversions are you losing because users went from a generic to a competitor query without a chance of seeing your ads?
- Is your brand favorability declining because of this strategy? I’ve conducted two studies around this topic over a three- to four-year span leveraging the Google Trend report and PPC data — and the answer was invariably, yes.
Sample of Google Trend Results
Date Range: January 2009 through January 2012
Goal: Gauge the search demand overtime around the ACME brand
Impression Share Report (also referred to as Share of Voice):
What is the percentage of impressions you are currently capturing out of the total impressions available in the market(s) you are targeting? More importantly, what is the percentage of ad impressions you may be losing? Once you know the volume of impressions you are currently leaving on the table, you can calculate the approximate volume of clicks and conversions based on your historical data.
Understanding the PPC opportunity will allow you to refine your media plan and answer some important questions:
- Is my current budget in line with market growth? The search landscape is anything but static — blink once and you might fall behind the curve. For instance, Adobe’s Q3 2012 Global Digital Advertising Update shows that search marketing spend in the U.S. grew 11% year over year. You might want to ask yourself, has my share of impression decreased or increased over time? Is the competition closing in, or am I driving the cost-per-click up by being too dominant?
- How do I want to position my company against the competition? Am I capturing enough share of impression to sustain this goal?
- Can I extrapolate the PPC data to determine the desirable budget across the other channels?
ROI-related metrics are usually the main focus of a search engine marketing (SEM) campaign, but monitoring the share of impressions is a good indicator of where you stand against the competition. The key to success is to find the right balance between achieving your ROI goals, staying competitive on the search result page (SERP), and maintaining your volume of conversions.
Note that Google provides the impression share at the campaign and ad group levels only. Inactive keywords are not part of the calculation in the AdWords interface, but your Google Account Manager can usually help paint a more “customized” picture. Alternatively, Bing and Yahoo! can also provide this type of data, but you might have to reach out directly to your account team instead of accessing it through the user interface.
Finding the Impression Share Data in AdWords:
Determining impression share in markets such as China, Russia, and Korea will require more work since Google is not the dominant search engine in these markets. First, look for Google’s most recent market share in each country, then apply the appropriate multiplier to calculate the full opportunity:
Example: The spend opportunity on Google China is $100,000 and Google represents 15% of the market share in this market; the calculation is as follows:
$100,000/0.15 = $666,667 total opportunity for China (assuming a similar average CPC)
Leverage SEM real-time capabilities to test the potential success of messaging through aggressive A/B testing. Protip: I always recommend to clients that they launch a PPC campaign a few months prior to a major product launch.
It might be truer for some verticals than others, but chances are the “word” is already out, so why not capitalize on the buzz without revealing too much? A recent Google “Launch Predictor Study” focused on the technology vertical found that 34% of searches happened before a product launch. Remember that your pay-per-click campaigns leave you in total control of the budget, messaging, and landing page environment. The upstream traffic you capture can be invaluable to fine-tuning your cross-channel initiatives.
Example: Your creative agency has produced three storyboards and you need to choose a winner before going into production for TV, billboards, radio, print:
Scenario #1: Price centric
Scenario #2: Product features focused
Scenario #3: Environmentally friendly (social responsibility)
You could use your gut feeling or you could leverage your PPC data to make an informed decision. Once you have decided to use PPC as a testing ground, it just becomes a planning exercise. First, understand the end goal (e.g., increase the quality of your reach vs. increasing frequency); and the questions you are trying to answer (e.g., which message resonates best with my audience(s)? Are consumers associating the product to the brand?); and then build a SEM strategy to help fill in the gaps.
At this point, you know how much you want to invest, where you want to invest it, and which message carries the most weight. But is your website ready to convert your highly qualified PPC traffic into sales, leads, or brand advocates? Whether your SEM program is mature and just needs a boost to reach to next the level of performance, or it’s in the infancy phase and you want to accelerate the learning curve, landing page optimization is critical to success. This is a complex topic deserving of its own post, but you need five things before getting started:
- Relevant user experience (search query to landing page)
- Consistent messaging from copy to call to action on the landing page
- Optimal design to maximize conversion rate
- Robust web analytics implementation (quantitative and qualitative evaluation of the traffic)
- Ongoing split A/B testing strategy (e.g., champion vs. challenger pages)
Once those five stars align, you can leverage your PPC campaigns to drive traffic to select pages, as well as track back-end metrics (also known as post-click metrics), such as: conversion rate, pageviews per visitor, time on site, bounce rate, average value of an order, etc. Each value point will tell a story and give you the ability to course correct or amplify your efforts.
Measuring the impact of offline activities is a challenge for advertisers, but website and PPC traffic data can provide invaluable information. Do you observe a correlation between offline investment and volume of related search queries? Is the conversion rate on your site higher during or just after a big offline media burst? You will need a robust web analytics implementation and enough PPC data to draw relevant observations, but the findings are well worth the effort.
Isn’t the ability to validate your offline spend gratifying? Why stop there? Let’s move on to social media signals and SEO. Over the last two years, advertisers have gone through several phases:
- Building a fan base
- Monetizing their fan base
- Retaining their fans
I think we can push the envelope and extract deeper knowledge — we are in the optimization business after all! The search query report in AdWords, for example, will give you information on how your audience is evolving.
- Are your keywords and ad copies up to date?
- Have the social signals influenced the user behavior? (e.g., which search platforms are most commonly used by your audience, what factors contribute to the decision process, etc.)
- Has brand exposure through social media impacted your key performance indicators (KPIs)? In the 2011 PPC Edition, MarketingSherpa reported the average click-through rate (CTR) doubled for organizations that had integrated their paid search and social media campaigns. Various case studies also show improvements in conversion rate.
The same questions and methodology can be applied to SEO/SEM integration. The reality is that most advertisers don’t have the budget to capture 100% of impressions on 100% of their core terms. Your PPC data and integrated reporting will help you build a cohesive strategy and guide you through the weeding process. Where should you allocate your paid media budget? You have limited SEO resources, what should you focus on? Based on your business and marketing objectives, the SEM numbers will point you to:
- High ROI terms
- High conversion terms (those could potentially yield a negative ROI)
- Brand awareness terms to build your upper funnel
Testing is fun. Looking at the performance trending upward, and collaborating with other agencies to improve the overarching program’s bottom line are exhilarating, but there are a few things to keep in mind:
- You need a statistically relevant set of data to make informed decisions. No matter how tempting it is to make tweaks on the fly, they will be detrimental if you try to go too fast.
- The higher the volume of data your campaigns produce, the more quickly you will be able to reach the end zone. In terms of conversions, a higher price point usually means fewer conversions (e.g., selling a CRM vs. a pair of socks) — it will take more time to collect sufficient data and optimize.
- Data is volatile and metrics directly impact one another. One change will cascade–sometimes as expected, sometimes not.
- Data is great, but data can also be overwhelming. Start small.
- If not formatted properly and applied with context, the data will not tell you the real story and will be meaningless.
Pay-per-click is a powerful tool for both online and offline marketers. I have worked with a lot of agencies – big and small – throughout the years, and one of the challenges was to not only share the data, but make sense of it. I have seen 30MB Excel spreadsheets (also known as “wall of data”), but without context, without collaboration between the campaign managers, that data is not actionable.
I joined BlueGlass in August to tackle just that — how do we truly integrate the SEO, social media, and paid media pieces not just from a campaign management perspective, but from a reporting perspective? How can we help advertisers maximize their offline and online spend by leveraging our data? It seems easy on paper, but collecting and digesting the data into insightful and actionable recommendations is not as simple as you may think. As I said earlier, there is more than meets the eye.