I once moderated a focus group for Google where we split the participants into two groups: small agencies and large agencies. One key difference I saw between the groups was their willingness to share best practices.
The small agencies seemed reluctant to share for fear of giving their competition an edge. The larger and more successful ones were openly sharing because they thought that today’s big insight would be old news soon, so they might as well share it and build up their own credibility in the process.
In that vein, as another year comes to an end, I’d like to share some of the things I’ve learned in 2014 in the hopes that you might benefit from it in your campaigns.
Chatter about account automation continues to grow in the industry. Two years ago, I was the only person at conferences speaking about the power of AdWords Scripts for automating rote tasks – today, there are entire sessions devoted to scripts, like the one I will be speaking on at SMX West in March.
It’s clear that advertisers realize some level of automation is required to remain successful in online marketing. We’ve certainly seen this at SalesX, where we went from spending a lot of time manually doing routine maintenance, to leveraging automation monitored by account managers to free up their time to get creative. With the time they’ve regained, they’ve moved beyond the basics of account management and developed new strategies to move accounts forward.
I’ll share some of those insights later, but the larger point is that when you hire smart people, you should enable them to be creative, and not waste their limited brain cycles on menial tasks that can easily be automated.
In the process of creating automation for SalesX, we learned that managing scripts comes with its own set of challenges, especially when dealing with many accounts. That’s why earlier this year, Optmyzr created Enhanced Scripts for AdWords, a way for non programmers to customize sophisticated scripts.
Even account managers without any programming skills are able to take advantage of some of our best automations, like automatically creating an entire campaign using data in a Google Spreadsheet.
If you decide to write your own automations, some of the complications you may encounter will be related to timeouts, handling more than 50 accounts (a limit Google imposes), maintaining versions and upgrading scripts, and running the same script with different settings for different accounts.
I’ve covered how to get around some of these AdWords script limitations in previous posts.
Search engine marketing is great because it is data-driven. Unfortunately, many keywords simply don’t have enough data to make any statistically sound decisions.
It may be that the account doesn’t get enough conversions, or the data may be spread too thin across many campaigns. So what do you do when you think you don’t have enough data to make a good decision?
One trick is to measure more than a single conversion, for example you could measure both micro and macro conversions. A micro-conversion is a desirable action the user took on your site, but not the ultimate conversion goal. Because micro conversions tend to be a precursor to the macro conversion, there is usually more data to work with.
There are a plethora of things you can do once you have more data (like that from Google Analytics) available in AdWords. In fact, I devoted an entire column to Google Analytics reporting for AdWords this earlier this year.
At Optmyzr, we noticed that it was hard for advertisers with a lot of different conversion types to see their data at a glance, so we integrated Google Analytics data into both reporting and our MCC dashboard. Now, advertisers can quickly see how many macro and micro conversions each account gets. So even if the data for the ultimate conversion goal are sparse, we can still show meaningful data.
One of our account managers, Emmanuel, came up with a micro conversion within AdWords which he calls cost-per-engaged-click (CPEC), a calculated metric that helps him optimize accounts.
By looking at Google Analytics data like time on site and bounce rate, he’s able to determine which clicks lead to engaged users. He can then spend more for those clicks because they’re more likely to turn into a macro conversion in the long run.
Having this metric available helps tremendously with bid management and keyword selection for accounts that are still growing towards a point where they have enough macro conversion data.
The introduction of the new shopping campaign type in AdWords earlier this year has been both a blessing and a curse to account managers everywhere.
It’s a blessing because it’s more logically structured and easier to set up, and the ads get much more prominence on the results pages, making this a big opportunity to grow conversions.
It’s a curse because the AdWords interface is woefully inadequate to manage bids for an account of even a modest size, and the ease of setup has opened the floodgates to new competitors.
At Optmyzr, we immediately knew no account manager should suffer through setting shopping bids in the AdWords interface, so we built a tool for managing shopping bids quickly. And when we say quick, we mean it — one customer went from spending 5 hours to 15 minutes to manage the same account.
What we have not yet built a tool for, and what remains challenging, is setting up a feed. For account managers accustomed to working with keywords and ad texts, the transition to managing ads through data feeds can be daunting.
Because the data in the feed are responsible for both the text of the ads and their eligibility to appear for certain queries, it is critical to have a well-managed feed. While I don’t have any great tips to make this easier, just know that you should allocate some serious time to feed setup and optimization to remain relevant in this growing segment of AdWords.
Many clients still question the effectiveness of running ads on their own brand keywords. Our findings mimic those of Google, and we noticed that running brand ads is an effective use of budget. For clients who have long sales cycles before a conversion happens, the brand campaign is usually the one that ultimately captures most of the sales. By the time users have gone through their long decision process, the thing they remember most is the brand. Hence not advertising on your own brand opens the door for a competitor to get a foot in the door right at the time when one of your prospects is ready to convert.
For advertisers who have a very strong brand already and who see lots of SEO traffic from their own name, we use Google analytics to compare the amount of organic traffic lost with the increase due to AdWords. It’s especially telling when you also include conversion value data in the report because it shows you exactly how much extra revenue was generated (revenue gained in SEM – revenue lost in the SEO channel – cost of SEM = benefit of doing SEM and SEO together).
One thing we’ve known for a long time but validated once again this year is that we should report brand campaign results separately from everything else. Brand campaigns typically have higher conversion rates and lower CPCs, so if those stats are mixed in with the entire account, it can present an incorrect picture of the health of an account. You should also ask your agency to report separately on different campaign types, e.g. shopping ads vs. brand ads vs. remarketing ads, etc.
In addition to the work we do managing AdWords accounts and building tools, there is also a lot of work that is done managing client expectations. An agency’s job is to guide clients, not simply do everything they ask. Especially when working with local businesses, we find we need to do a lot more coaching of clients because they don’t understand SEM very well. In one case, a client had conflicting goals, and after we met his cost per acquisition goal, he was unhappy about the average cost per click being too high.
It can be difficult for a smaller business to stomach a $30 cost per click or be happy with a Quality Score of 5; however, in the grand scheme of things, CPC is just a lever to help achieve a target CPA or ROAS, and Quality Score is just an attribute Google provides to help identify areas of the account that have a weak relevance. When you work with your stakeholders, it’s important to push back when you get requests that don’t make sense because if you let someone with no AdWords experience get involved with the tactics of the account, things can go south very quickly.
I also see a lot of people who expect that AdWords is a magic bullet that can kickstart or revive a flailing company. The truth is that AdWords can’t cure a broken business model. If sales aren’t happening offline, chances are you won’t get very many sales online either. And even when you do, the inefficiencies of a broken business model may make it extremely difficult to achieve profitability in SEM where you are sure to be competing against at least a few more efficient companies.
Those are a handful of the things I learned in 2014. Online marketing is such a fast evolving field that nobody can expect to be perfect all the time. I hope that some of the things that took me some time to learn will help you be more successful in 2015.