All of the content in the Paid Search Horror Stories series was presented by our own Senior Account Manager Donna Lagow at Digital Summit Portland in July 2017. Here’s a complete video of Donna’s Paid Search Horror Stories presentation.


A big round of applause, talking about paid search, Donna Lagow.

Thank you guys for joining us this evening as we wrap up, especially as we’re leading into the very important happy hour part.

What our session is about today is specifically focused on some of the large perils and disastrous performance that we’ve seen in PPC account management. So I’m going to be kind of pulling back the curtains on what we’ve seen as an agency, and hopefully you can learn from some of the lessons that we’re going to be covering today and how to not fall into those same pitfalls of money loss, really.

Before we get into that I’ll tell you a little bit about me. Again, my name is Donna Lagow. I’m the Senior Account Manager at (un)Common Logic (formerly named 360Partners) in Austin, Texas. We are a boutique agency. I manage clients across the PPC, SEO, and CRO vertical, and my clients actually span across a myriad of industries, from web hosting clients to the financial services fortune 500 group, as well as large E-com clients. So lots of, again, pitfalls of performance to be able to pull from examples here.

Before we get started, I also want to tell you a little bit about Austin, Texas. This is where we’re from. If you guys have not been there yet, you should. It is the queso capital of the nation. I live there with my Great Dane Cecelia. She is quite great, all 120 pounds of her. And I really like spooky things, which aligns very well with the topic here today.

So without further ado, let’s get into the good stuff, or the disastrous stuff, as I’ve been calling it. We’re going to be going through three of the primary storylines today.

  • First, how meeting our members overall was actually killing the company.
  • Next up, the rise of the scripts, or the rise of the machines, as I call it, and how that strategy nearly slaughtered our ROI.
  • And then last but not least, three fatal errors that we see when going through paid search audits as an agency.
How Meeting Our Numbers Was Actually Killing the Company

So: how meeting our numbers was actually killing the company. How’s it even possible? you ask. That person wants to know.

Alignment: it’s key. It’s really important. When you’re hearing that word, it’s one that we use in our day-to-day quite often. It’s usually in reference to a goal: are we aligned in terms of the reach goal that we’re aiming for? Are we aligned in terms of the new business shift and product target that we’re trying to meet? It’s pretty important to get that all established up-front as part of our vernacular. Alignment’s really just how we feel, relative to a concept, I guess.

So, we had a client—this is one of our larger online clients as well—and they are in the in the business of selling large basically storage containers. Those mobile mini office things that you see alongside railroads.

And when starting to have initial meetings with them and onboarding them as a client, their senior management team walks us through and articulated their goals for us. And they took us through, overall, what their primary acquisition focus was, Again, we’re going to be focused on mobile offices, specifically. A secondary and far less important goal for them was this storage containers category. And they said, “Really, I don’t want to allocate any spend at all whatsoever to the PODS category; that’s a money-waster for us.”

So when looking at the account, this is basically what we expected to see: beaucoups of money allocated to mobile office. Well, some spend towards storage containers. Obviously, as much as we can get for brand. And then zero’s been allocated to the PODS category, just as senior management directed.

So what we did was started our account analysis—and, really, anyone can do this. We pulled long-term search query report data and basically went through and themed each of the queries based on the product category that it was associated with. Then PivotTable’d and aggregated that data to get to spend number.

So, again, this is what we would expect to find. Where in reality, what we found was something much different.

They were spending only about half of their budget on their primary product offering that they wanted to grow; much less on the storage containers; still doing about right on brand. And a whopping 30 percent on the PODS category that they did not want to put any investment into it all whatsoever; a loss leader for them.

How could this happen? It goes back to alignment. So the marketing team was given a series of goals. The goals were:

  • Get me 2,500 leads; I want to take them to sales
  • I want them to be about $40 a lead, and
  • Our budget is going to be around 100k

Great! So when looking at the goals… Okay, leads right on the money, bullseye. Our cost per lead, fairly close as well; I’m comfortable with that. And then our spend—again, right there on target. So overall we’re winning, right? We’re hitting goal, I feel good about it.

But should I? Not really. Overall, what this strategy was doing was incentivizing performance to the actual goal target, not towards the product line and the product growth is what the senior management team wanted.

So what you have to do with your marketing teams is: be very, very clear in your budget-setting process and very clear in terms of how they are segmented and how they’re managed—there should have likely been a separate specific budget for the overall storage category itself. And then you need to align your spin with those goals as well, and then make sure that you’re doing those checks.

Again, the very simple process of going through and reviewing your query data to inform whether or not the actual business goals—the company goals that you want to hit—are how you’re allocating your dollars.

And like all good horror stories, this one has a sequel. Dun dun duuun..

The Menace of Brand Keywords

When our marketing team, again, looks as if they’re winning at performance hitting goals, just as we illustrated, but really they’re killing our new customer base. How can that be happening?

One of the stories that comes to mind for me is, again, in a new customer conversation, we were talking to the CMO about their overall performance, and of course goals came up and how they’re pacing against those.

So immediately the conversation was: “We have a 10% conversion rate”—this is a large e-com luxury online retailer, so… 10% conversion rate—and they cited an astronomical return on ad spend.

And our response was: “I am so sorry.” How could that be? You should not have that high of a conversion rate if you’re actually a large e-com retailer. It’s almost impossible.

So what that signaled to us was, in general, that they were targeting a way too small demographic and really not focused on growing the business and scaling. So what we then saw in their actual numbers was that they were focusing heavily on their “profitable brand terms.” And, again, I think everyone here knows that that can lead you to a very quick march to demise.

So overall we started to see this curve of their new customer demographic continue to dwindle and dwindle and dwindle, given that incentivized focus on that higher ROAS target and, again, the lower focus on new customer acquisition.

In general, the stat that you will hear often is: for new customers, it costs about 4 to 10 times as much to actually acquire that customer. It’s a heavy investment. You’re introducing yourself, basically; it’s a hard sell. You’re going to have to introduce yourself multiple times. And it’s really essential for growth. Obviously, the more new customers you get, the more existing customers that you retain. It all makes sense, right?

Folding back to the other story, in terms of how you map back to the goal, you have to segment them out. A lot of the way that we manage our goals in our agency is very heavily acquisition- versus retention-focused.

We have very specific targets in terms of the return, in terms of the conversion rates, in terms of the investment across the board, in terms of the CPCs as well. And you have to set up your spend allocation accordingly.

Overall, when you’re mapping back to the types of goals that we said, you’re really not taking into account that greater difficulty in acquiring a new customer. And the balancing scales that that lower cost of acquisition has for you on a current customer side is the thing that can be a threat overall to performance.

It’s really incentivizing marketing teams, without even realizing it, to preclude the new customer demographics. If you have a blended CPL or CPA without clear and illustrated goals for each of those, you’re likely driving your overall performance in an efficiency model versus a growth model.

How do you elude that? Overall, again, different budgets, different goals, very easy. and at overall investment in the new customer drivers. Things like display, things like high-funnel keywords. The more money that you put there, the more long-term customers you’re going to see on the retention side as well.

And, in general, when you’re setting up goals strategically across the two types of customer demographics, you always have to measure them and segment in that way as well, so new customer campaigns can’t be evaluated on the same standards and goals that the existing customer campaigns are.

I think we’ve got that point down. Next up: the rise of the scripts—of the machines, I call them.

Rise of the Scripts: The Strategy That Nearly Slaughtered Our Revenue

Who doesn’t love scripts? We all do, I’m assuming. They make our lives much easier. They can scale quickly, they take into account a myriad of conversion touch points to inform the automated decisions that they’re making. So, you know, what’s wrong? What’s the problem?

It’s really our brains and the fact that we’re stopping using them. Overall, when people implement scripts, oftentimes what we’ve seen is an over-reliance on the technology and, really, the removal of those periodic checks that you do in the manual format generally,

The over-reliance on scripts: it’s so convenient! They’re built right into the platform! They give the illusion of management and efficiency but, again, you have to be able to go through and check that how the tools are actually performing, or in line with the overall product goals that you’re setting.

Who’s really in charge? That’s the biggest question. The thing that you have to look to, again, is the manual review of those specific tools. So your script and your tools are going to be as good as the person behind them, right?

You have to know your accounts, you have to set up your scripts in line with the overall budget goals and strategy of the company, and then you have to check them and check them often.

A thing that we’ve seen quite often in reviewing script implementations is what I would call an erroneous layering approach. Some of the scripts that I’ve seen—and within bidding functionality, you can do things like “I want to manage my bid to this target CPA, I want to also hit this average position, I also want to hit this max CPC.”

Well, what happens when you continue to layer and layer? What’s really driving performance? How do you know? It’s a compounded impact, really, and what it leads to overall is a heavy reliance and a removal, as we call it, of performance.

One of our clients—this is actually a recent story—they had to take their actual ad spend back in-house. They were with an agency for probably four years, that being us (pardon me), and they decided they needed to get more efficient with their spend.

So they brought it back in-house, and with their smaller team, they were looking for ways to get more efficient overall, as well as internally. Hence, scripts.

So scripts were applied for bid optimization, scripts were applied for budget management. What we saw over time and evaluating their account—because they actually ended up coming back to our agency and asking us to do an audit—when performance continued to decline, was there were scripts that were set to turn keywords and ad groups on and off based on performance.

And the performance metrics that they were allowing for, in terms of those thresholds, were pretty liberal. It wasn’t that the rule wasn’t actually fairly sound, it was that there were no periodic checks to go back and review what had been turned off, and evaluate that for things like larger performance swings.

Seasonality, for instance, was that being taken into account? Things weren’t being manually turned back on, so over time what we saw for them was a heavy degradation of their actual keyword volume. The keywords within their accounts that were active, as well as the keywords that were driving traffic overall, and the account continued to have less and less lead volume; less traffic, as you would expect; and results continuing to go down and down and ROI out the door.

So that was a primary recommendation that we’ve put forth to them: if you’re going to implement, you’ve got to make sure that you have the people power to make sure that the decisions that you’re making are the actual right ones.

That, again, points to the fact of doing the work. I think that’s really the primary message here: while scripts are great—we love them, we use them judiciously—it’s the quality and checks that you have to make sure you continue to interwove through your processes as well.

The Three Fatal Errors of a Paid Search Audit

Next… I like this one. I really like paid audits; I’m a geek with a spreadsheet. We do paid audits generally any time we’re onboarding new business or just really trying to help out a referral in terms of improving their performance.

So some of the big flaws that we see in paid search auditing are, in general, never doing them. Why in the world would you not want to understand what’s going on in your account at a very granular level? We’ve actually had clients say “I haven’t done an audit over the last year or two.” Eeek!

So when you’re doing that, you’re allowing for a status quo. You’re allowing performance to be the “good enough” phenomenon.

We, actually, on the (un)Common Logic (formerly named 360Partners) side, have what we call an account swap situation. We, probably every six months, will rotate accounts between our account teams and then evaluate how performance could be better. Again, is it hitting long-term growth goals? Are we efficient enough? Are the big things that are the money drivers for us as effective as they can be?

So, again, it’s an opportunity for you to get better. And who doesn’t want that?

In general, the things that you’re looking for: fast wins, you kind of want to follow the money; as well looking for opportunities to reduce wasted spend; and in general an opportunity to buffer up your strategy and your long-term game for your account.

Next up: the beast! You really want to focus on the beast, and the things that matter. Really, it’s—what that’s indicative of are the big huge things in your account that are:

  • The primary drivers of revenue
  • The primary drivers have spend
  • The primary drivers of orders

When you’re looking at those larger aggregates of spend, you’re likely going to be able to get to, in short, the problems, the areas that are absorbing that spend and how it can be improved better.

Focus on that not on the little guy. Basically, you can spend a lot of time focusing on something that’s accrued you ten dollars over the last two months.

In order to do this, you really need to also focus on what? Leads, and understanding that they’re really not all created equal. What does that mean? Well, it goes back to being able to focus, again, on the things that matter.

Where possible, you want to be focusing on lead-to-revenue ratio, your cost per sale: how do those things grow and be impacted by your overall account performance?

Optimizations on CPL can lead to what we call the “death spiral.” We’ve had a client who wanted to bring their CPL down. “I would like to take it down 10%.” Okay, that’s a fair goal. I’ll take that challenge. But what they weren’t focusing on, again, was that end-to-end measurement.

This was an online academic client and they weren’t looking at, really, the cost per enrollment. So in general, more of that lead-to-revenue ratio was what they were removing from the equation.

So when you actually flipped the numbers and focused on the end-to-end tracking and performance versus only CPL, it was a different story completely. And, again, this was a heavy impact to their performance and how they continued to actually move up from there.

Next, this one. Why is last-click still a thing? Why, why, why? I know that this is something that we talk about often, but it really can’t be the last point in terms of your auditing process.

So what does that mean? You have to think multi-channel; all the attribution platforms and really even what we can find in Adwords now gives us the ability to look beyond the last click.

We can evaluate things like assisted conversions, cross-device conversions, anything that’s contributing overall to the myriad or numerous touch points that a customer may have before they actually reach that culminating last click.

Evaluating all that basically helps you look to what’s contributing at the higher funnel level and not putting off the top of your funnel. In short, that’s a classic error that we always see: “I need to get more efficient, I need to cut out 20% of spend, so I’m going to cut out everything that’s over a certain level in my CPL.”

Well, without looking across the board at what else was contributing in terms of traffic, maybe that was your second and third touchpoint to that conversion. Again, you’re creating a gap in your conversion process overall.

And lastly, you can’t preclude the website itself. So while we like to see what’s going on within performance on the PPC side, we’re obviously very interested in landing page performance as well.

But what happens after that? What happens after they start the conversion process, as they’re moving through the funnel? Are there opportunities to improve that? Is there friction that’s being created that can be taken down?

You can’t just stop at the last click on the site, either. In short, you have to keep moving all the way through the full customer journey and the experience to look for opportunities to improve the customer path. That was a big one.

I think that is all for the horrific stories that we’ve seen on our end. I’d be very curious to hear some of yours and definitely, please reach out. I like hearing these types of interesting quirky stories from PPC accounts, so please get in touch with me; I would love to hear from you.