As with most facets of marketing, PPC budgets tend to fluctuate with the seasonality of the product or service they’re supporting. Short of guessing or using rules of thumb, how can a digital marketer create PPC budgets that are based on metrics, rather than hunches?

If year-over-year performance data is available, turning to those numbers would be a helpful starting point. However, while historical data can provide general direction, it can also be misleading. Your daily budget, max CPC, quality scores, and ad pacing and scheduling can all affect the performance of a campaign and cloud your judgement as it relates to the number of people searching for a term.

Ideally, the digital marketer will be able to arrive at a monthly budget that relates to the seasonal fluctuations in search interest in order to optimize allotments over the course of the year.

PPC budgeting solution

Using Google’s Insights for Search tool, online marketers can base monthly budget allotments on actual search volume data Google.

See the following chart for the term “baseball gloves.”

While the chart doesn’t provide actual number of searches, it does provide normalized values of searches, meaning that it’s easy to discern relative change in search interest over time.

So how do these numbers affect your monthly budget?

In Excel, enter each month’s number into a column. In this example, January is 47, February is 60, March is 84, and so on. Once you’ve entered all 12 months, average the values.

To find the budget index, divide each month’s number by the 12-month average. The result for each month provides with you the percentage above or below the average for the search term.

Finally, multiply  your monthly budget by each month’s index value. If your monthly budget is $1,600 and January’s index is 1.05, your actual budget will be $1,680 ($16,00 x 1.05). If interest falls sharply in February to 0.70, your budget for that month would be $1,200. Once you calculate all 12 months, your yearly budget should remain unchanged, but each month will be able to fluctuate based on  actual search volumes.

And it’s that simple. Your PPC budgets now rise and fall with seasonal interest, based on real data.

Pic by Josh Clayton.

There’s a good chance you or your clients are tweeting and Facebook posting links to blog articles, press releases, and new sites, but how do you know what’s effective and what’s floundering? You can keep track of your social media marketing efforts and track fairly granular inbound referral information using a set of free tools, which is great.

Here’s a quick tutorial on how to setup your social media tracking and analytics system.

Required tools

• Google Analytics account: if you don’t have one, sign up for an Analytics account. But what web analytics pro doesn’t have a GA account, right?
• Google URL Builder: in conjunction with Analytics, the URL Builder allows you to gain quick insight into the performance of your social media links.
• Spreadsheet program: such as Excel or Google Docs spreadsheet.

Social Media Tracking Method (the short version)

1. Determine the target page on your website for your social media campaign
2. Plan which social networks you will target
3. Enter your URL into the URL Builder, entering the necessary parameters
4. Record the parameters in your spreadsheet for reference
5. Link your content on the social networks
6. Track in Google Analytics
7. Adjust your tactics

Social Media Tracking Method (the full version)

Step 1: Let’s say your client sells hand-made t-shirts in-store as well as online. Your marketing strategy revolves around a “hot deals” page on your website that offer “buy now” opportunities for your 6 top-selling designs. All inbound activity will be directed to this landing page.

Step 2: Your client has a decent following on Twitter and and active fan group of Facebook, so they will be the primary target of your social media marketing activities. You also post images of your products on Flickr, so you’ll track inbound links from there as well.

Step 3: a guide to the URL Builder

• We’ll start with Twitter
  

  Screenshot of the URL Builder (click for larger)

• Enter the base address (e.g. www.shirtstore.com/hotdeals.php) into the “Website URL” box
• “Campaign Source” = the site where you’ll post this particular link, so for the Twitter links, enter “twitter”
• “Campaign Medium” = the kind of link you’ll be using, like a banner add, text link, email campaign, etc.
• “Campaign Content” = used to differentiate different kinds of content pointing to the same landing page
  • The Twitter component of this online marketing campaign will have two sources:
    • Post links, i.e. links you’ll post in your feed
    • Bio link, i.e. the link in your profile information
  • You’ll give them both unique names for tracking purposes
  • Since we can only build one URL at a time, we’ll do to the post links first, so enter “post_link” into this box (you can’t use spaces)
• “Campaign Name” = used to identify the larger campaign this link is a part of. Since your marketing efforts are focused on the Hot Deals landing page, enter “hot_deals_fall_09″
• Finally, click “Generate URL,” and your URL is built!

You’ll have to go through this process for each different link on your social media properties. So two for Twitter, changing the Campaign Content field for each kind, and for each Facebook link and Flickr link you’ll post.

Step 4: As you can see, this analytics systems results in a large inventory of URLs. To keep track of what values you’ve used for each tracking parameter, setup a spreadsheet that store each value used. Use the furthest-right column to leave yourself notes about how each URL was used, where and when. Again, the initial setup is a bit time consuming, but your efforts will provide solid social media analytics so you can focus your marketing efforts on activities that bring results.

Step 5: Now that you have your list of URLs in your handy spreadsheet, implement them on your target social media sites.

Step 6: Once your links are in the wild, you can move on to the exciting part: tracking the campaign results. Login to your Google Analytics account and click on “Traffic Sources” in the navigation on the left. The two sub-sections of interest in our case are “Campaigns” and “Ad Versions.” These reports will quickly overview the progress of your campaigns and the versions of your links you’ve posted on the social networks.

Additionally, you can view the Campaigns based on the other dimensions you set when building the URL, namely source, medium, and ad content.

Step 7: Watch your campaign for the next days and weeks to see which links are driving traffic to your website. Maybe you find that posting image links in your Facebook status drives twice the traffic as the Flickr photos, so you spend more time on that approach. Any combination of results is possible, but without analytics and metrics, you are completely in the dark about the effectiveness of your small business social media marketing efforts.

Hopefully this tutorial to setting up your social media analytics system has been helpful. Please comment if you have any other helpful social media marketing tips!

Pic by tricky ™

One of the most powerful facets of web analytics is the absolute ability measure everything. This ability creates a certain level of accountability for us, our managers, and our clients as we set goals for the coming year. Vague goals are neither helpful nor needed in light of the objective nature of web analytics.

As we approach 2011, I’d like to remind us of the notion of SMART goals and objectives as a rubric for measuring the efficacy of the goals we set for our companies and clients. When setting your web analytics goals for 2011, each goal should be:

• Specific – Is the objective clearly defined? Will key players grasp it easily?
• Measurable – Can you track progress? Do you know when you’ve reached it?
• Attainable – Is this a realistic objective? Is it within the project’s scope?
• Relevant – Does it contribute to attaining the goal?
• Time-bound – Have you set a deadline? (note that some people use “timely” for T, but timely has a slightly different meaning than time-bound)

Rather than setting the unspecific objective of “increasing hits to our website,” you’d be better off with something SMART, like this:

• to improve average monthly unique visitors by 15% by the end of the second quarter of 2011, or
• to increase the conversion rate on LCD TV product pages by 8% be March of 2011

Some have added another two points to the SMART planning system:

• Evaluate
• Revaluate

These last two points encourage us to strive for continual improvement and accountability of our objective setting. May 2011 be your SMARTest year yet.

Pic by jeff_golden

Warning: this is  rant; please read no further if you abhor strong opinion

The “E-blast” is an apparent favorite tool among online marketers. The notion of mass emailing thousands of mildly interested users pieces of promotional information is too tempting to pass-up.

But does anyone want to be blasted? To have their inboxes filled with explosive material from marketers too lazy to focus their list? In 2010, with all the tools of segmenting and list building, why on earth would anyone ever implement an “E-blast”?

I’ve heard clients ask to use “E-blasts” as part of their plan, and even more disappointing, I’ve seen the term “E-blast” as an official offering of big-time interactive media companies.

It’s 2010 and time to retire this inaccurate and violent terminology. If you’re blasting, you’re doing it wrong. Segment your opt-in list, please, and use email marketing to strengthen existing relationships, not to desperately try to build new ones.

Pic by ve0_

What do PPC benchmarks and Santa Claus have in common? They don’t exist.

I realize this may come as a shock to many, but the existence of a universal all-encompassing PPC benchmark is more implausible than a charitable and magical elf and his flying reindeer.

We can all remember believing in Santa, and I’m sure we can recall really, really wanting to believe.

Santa Claus provided a no-nonsense  system:  behave and get presents; misbehave and get coal. No gray area.

Benchmarks provide similar comfort for marketers, especially for those not entirely comfortable with the digital realm. Someone may not totally understand what AdWords does, but at the end of the day, if he can show his boss that last quarter’s campaign outperformed the benchmark, then he’s been a good boy.

Thing is, life is complicated and anything but black-and-white. Life is messy and complex and gifts don’t magically fall from the sky. Sadly enough. Similarly, despite ardent desire for a PPC benchmark, we’d be naive to think that campaign measurement can be so cut-and-dry.

Granted, there are some arbitrary numbers floating around regarding the CTR of an AdWords ad, but the fact is that the number of variables involved in a PPC campaign violently undermines the viability of a benchmark.  Consider a few of the variables that change from campaign to campign:

• keyword buy
• keyword match type
• daily budget
• max CPC
• geographic targeting
• scheduling
• time of the year
• ad rotation and frequency capping
• ad copy
• the inclusion of product images or additional site links
• competitors’ placements
• the type of product or service being advertised
• landing page copy, images, calls-to-action, form length, among others

The value of a benchmark is in it’s apple-to-apples comparison, but with some many variables involved, such comparison is neither possible nor desirable.

Now that You Know the Truth

In spite of this well-reasoned rant, your clients and bosses will still ask for benchmarks to help them measure the efficacy of a PPC spend, and you’d better have an answer that makes sense.

Here are a few things I recommend to move the discussion beyond arbitrary benchmarks:

• compare inbound traffic – how does PPC traffic compare to other channels (e.g. direct, referral, organic) in terms of engagement? Maybe your PPC traffic converts at a higher rate than all other channels. Good job!
• compare PPC traffic to itself – if you’ve run a PPC campaign for several months or at different times over the course of the year, comparing historical performance can help you understand the effectiveness of different combinations at different times of the year.
• educate stakeholders – PPC, as with digital advertising in general, is a brave new world for many decision makers. Our primary role, I think, is to help lower the bar for digital by helping others understand what we do. Talk them through the concepts in this article to help them understand that a benchmark in this case is neither realistic or helpful.

I know, discovering the truth can be painful, but it’s also formative. Armed with this awareness, I hope to see us transcend make believe benchmarks and to help understand that power and complexity of PPC campaign measurement.

Pic by Grzegorz Łobiński.

Strategy comes first, right? Without a strategy, how will we know how to proceed? How will we get their without our map? Won’t we get lost? I’m getting scared already!

As is typical with most any advertising planning process, whether online or off, we begin with research and discovery, build a solid strategy, and only then do we execute. This workflow is true for social media marketing as well. In our blog posts and email newsletters, we always recommend that a group develop a sustainable social promotion strategy, and only then should they actually ”do” social media.

Before we continue, I should point out that these thoughts apply primarily to local small businesses, and not to enterprises or otherwise large companies that have dedicated staff to maintain their social presence.

The key difference is that small business owners are often the ones responsible for their business’s social media execution. If not, the duty usually falls on one of their employees who is responsible for covering the phones, dealing with customers, merchandising the floor, and somehow amidst the chaos finding the time to tweet.

It is this situation, which is less than ideal but all too common, that calls for a revised perspective on the social media process.

Who needs a map?

When we recommend the development of a social media strategy to this kind of business, have we truly considered the nature of their situation? Telling a small business owner to develop a comprehensive social strategy is like telling a person to use a map to find their destination without asking them if they even know how to turn on the car.

Taken another step, maybe they don’t want to go on the trip at all. But they must have a map!

That said, I think the most plausible and effective social strategy for SMBs and other groups who are only just learning to turn the key is to, as Nike once said, just do it.

We should advise them to create a Twitter and Facebook account, start connecting with people, sharing information, and building interest around what they’re doing. Sure, we can provide practical tips, like teaching them what “RT” means and how to use it effectively, but at this stage, elaborate strategies have no place.

Let them get behind the wheel, and after a month or so, if they’re posting regularly and enjoying the process, initiate the roadmap discussion and help them understand how social media fits into the larger picture of online marketing. At this point, a discussion of strategy will have actual meaning, since the person has had several weeks to learn about the social ecosystem and get a feel for how everything works. You’ll be talking strategy with a fellow practitioner and not with a person who’s only heard of Twitter on the news.

So I say let’s avoid lengthy strategy meetings and discussion of how we can target influencers with engaging assets to act as platforms for conversation, and simply enable these people to learn the craft by getting their hands dirty.

Pic by Norman B. Leventhal Map Center at the BPL

“Light bulbs weren’t invented by exploring candles.”

If I were to be completely honest, I might be tempted to claim that despite all the technological trappings of digital advertising, the developmental nut of building a digital campaign hasn’t fallen very far from the tree of traditional advertising.

We assess the business need of our clients, develop a research-based plan, present to the client, then build and track the campaign.  Were he to observe the process, Don Draper would point-out that things haven’t changed much since the glory days of Mad Ave, save for the substitution of printed creative with digital creative.

In essence, there is  a one-to-one correlation between traditional agency processes and digital ones. The existing agency model has, in most cases, simply been ported to the digital realm. It’s true that in shops where traditional and new media coexist, the situation is difficult to avoid, but this post is a call for more intentional thinking surrounding the digital agency process.

What is the candle that holds our collective gaze?

What if digital admen and women looked past the existing models in search of a truly innovative agency model? Sure, regardless of the medium, traditional, new, or otherwise, agencies exist to enable their client companies. But the digital side is fundamentally different than the traditional aspects of advertising, particularly as a function of the interactive nature of the digital realm.

A digital campaign begins with an idea, as with a traditional campaign, but with digital, the campaign doesn’t end with the brand speaking to the audience through a bullhorn. Digital channels, as we all know, enable multi-directional communication between brands and communities as well as among and within communities. We create and utilize tools that facilitate and create discussion, and the content of our campaigns morphs in real-time with the arc of those discussions.

Digital practitioners exist on the cutting edge of human communication, but I think our general self-understanding ignores the revolutionary nature of our medium.

We fail to innovate because we’re focussed on candles.

I don’t intend to provide a methodology for digital agency conceptual innovation (yet…), but rather to plant the seed of dissent in the minds of fellow practitioners. After all, we’re already doing great work that our clients love. That said, digital people are among the smartest and most creative folks I know, so with a bit of hard work, we stand to move from producing great work to delivering true digital innovation to our clients and their audience.

The quote that opened this post is from a book called IdeaSpotting: How to Find Your Next Great Idea, by Sam Harrison. It’s an easy read packed with inspiring thoughts and quotes, some of which provided the conceptual spark for this post. You should check it out.

Pic by Peter Becker.

One challenge web analysts face is presenting enough data to make a clear point without overwhelming the audience. Web analytics presentations are often presented in PowerPoint presentations, thus challenging the web analyst to communicate information quickly and clearly.

Comparing time on page

Assume you’d like to compare the average time on site for your top Services page along with a few other contextual metrics. Your chart may look something like the one to the left.

This is decent information, but it provides no context for comparison. You leave your audience with the worst possible question: “So what?”

Without meaningful context, as we’ve previously discussed, your audience will not be able to discern the significance of what you’re presenting and will wander-off into questions that detract from your story.

The good news is that with a few additional columns, your charts can simultaneously provide contextual information regarding each page’s performance while allowing for easy comparison of each page’s relative performance.

Indexing your web analytics metrics

To beef-up your reports, you’ll add 3 columns to the existing table: the average time on page for pages in that section of the site, the delta (or difference)  between each page and the average, and the indexed performance of each page.

Indexes are used in various business settings, but in the case of web analytics, indexes provide a way to quickly compare the performance of a page among a set of pages.  When you’ve added the additional columns, your improved chart will look like this:

Improved time on page comparison

It’s now possible for your audience to get a sense for how each page compares to the section average (site average can work well too, depending on your site’s structure). The difference between each page’s performance and the average is also apparent in the delta column. Finally, the indexed performance, coupled with the happy/sad faces, provide a digestible overview of how the pages are performing.

To calculate the index, divide the delta of each page by the average.

Adding the additional columns to your report shouldn’t take much additional time, since most of your base numbers will export cleanly from Google Analytics. Also, if you’re having trouble formatting times in Excel, see my post about dealing with time formats in Excel.

Finally, I would be remiss not to mention that the inspriation for this post came from Avinash Kaushik’s Web Analytics 2.0, which is a must-read for anyone in the field of web analytics.

Pic by Migraine Chick.

A text without a context is a pretext for a subtext.

An old professor of mine used to repeat this truism at least weekly when discussing literary criticism, particularly with regard to the New Testament. His point was that one can make a text say anything, regardless of actual meaning, so a thorough understanding of the surrounding situation is necessary for accurate interpretation.

In the world of web analytics, the need for context is nearly as paramount. 

How many times of you seen a chart similar to the one on the left? At first it seems great. Lot’s of “key metrics,” right?

Not so good

But really, what can anyone tell from these numbers? I can tell you that the only discussion a chart like will initiate will involve people saying things like “Well, 32,162 seems like a solid number. What happened last quarter? Last year?”

And then everyone at the meeting will either try to remember or will simply ask you, the web analyst, point blank about historical performance. And no one learns anything.

Setting the stage for meaningful discussion

As a web analyst, one of your primary roles is to foster meaningful discussion in which decisions makers quickly understand what you’re presenting and ask questions that actually have meaning.

There are a number of ways to create this kind of environment, but one of the easiest ways is to always present data within context. Stop wasting everyone’s time with numbers in a vacuum.

Here’s the same data from the previous chart, but with additional context.

Now we can get a sense of the historical meaning of the current quarter. Your clients won’t get bogged-down is silly discussion about what might or might not have happened. The context is crystal clear.

Now, within context, the discussion can revolve around the situation: the fact that volume is up (visits), but engagement (pages/visit, bounce rate, time on site) is slightly down. Conversion rate is also down. So what next? What, dear web analyst, are you recommendations? How can we improve?

Adding this kind of information to your reports is quick and easy, but your constituents will be extremely grateful. Additionally, you’ll find that decision makers will begin looking to you for guidance. Whether  you’re at an agency or in a marketing department, key stakeholders will begin seeking your advice and asking great questions because, through data’s context, you fostered meaningful discussion about their digital efforts.

Pic by Luis Argerich

Every two days now we create as much information as we did from the dawn of civilization up until 2003. . . . That’s something like five exabytes of data.

That’s according to Google CEO Eric Schmidt in this TechCrunch article. Five exabytes of data every day. To put that figure in context, an exabyte is equal to 1 billion gigabytes.

Crazy.

Digital advertisers and marketers are doubtless partially responsible for this proliferation of information. Microsites, landing pages, crowd-sourced campaigns, corporate blogs, and other promotional tools have helped churn out gigabytes of  client-serving stuff.

Digital imperative

Not that this practice is necessarily harmful for anyone, but I think there’s something of a moral imperative for those in digital advertising, and other online professions, to be good stewards of our roles as content generators and curators, as well as enablers (in terms of crow-sourced and social marketing sites).

Ironically, we’re our own worst enemy, in the sense that as we distribute a high volume of digital content, we crowd the very spaces we’re trying to speak into. Highly targeted and efficient content, then, is good for the target audience, as we speak only to those who are interested, and good for our collective clients as we avoid clouding the digital sky with ad-based pollution.