SEM for SaaS: The Good, The Bad, The Ugly (Video + Transcript)
SEM, or Search Engine Marketing, is important for many SaaS companies to attract leads and grow their business. How can companies properly execute their SEM strategy to maximize ROI? What should you do and what should you avoid?
Loretta Jones, VP of Marketing at Delighted, and David Rodnitzky, CEO at 3Q Digital take the Deep Dive stage to discuss which companies don’t need SEM, the importance of testing ad copy, which metrics you should be looking at, and the difference in SEM for SMBs and enterprise companies.
And if you haven’t heard: SaaStr Annual will be back in 2018, bigger and better than ever! Join 10,000 fellow founders, investors and execs for 3 days of unparalleled networking and epic learnings from SaaS legends like Eric Yuan, Tomasz Tunguz, Chris O’Neill, and Mikkel Svane. If you don’t have tickets, lock in Early Bird pricing today and bring your team from just $999! (All ticket prices go up December 1st.) Get tickets here.
Announcer: Please welcome to the “Deep Dive Stage,” whale enthusiast and VP of marketing at Delighted, Loretta Jones, and CEO at 3Q Digital, David Rodnitzky, also an enthusiast of marine life, especially eels.
Loretta Jones: Thank you all for coming. What David and I decided to do for our session was basically just go over some of the common questions that we get around SEM.
We have about seven questions, and we’re just going to answer the questions and have a conversation about it. The first question is, David, should all SaaS companies invest in SEM? What do you think about that?
David Rodnitzky: It’s a good question. I want to say that we have 25 minutes, and we both got speaking coaches, and they said speak very slowly. But because we have 25 minutes, and because we want to give you guys as much information as possible, we’re going to speak really quickly. Don’t blame us.
Should all companies invest in SEM? I run what’s primarily an SEM agency, so you’d think my answer is going to be yes, but I’m going to give you five reasons why you shouldn’t run SEM. Another way to say it is, “Here are five things you need as a SaaS company to be good at SEM.”
Number one is, people have to be searching for your product. If you’re selling CRM, great. People are searching. If you’re selling some newfangled price optimization Hadoop database that no one searches for, no one’s going to search. You’re wasting your time at SEM.
Number two, are your economics as good as your competitors? If Salesforce can pay $1,000 a lead, and you can pay $10 a lead, you’re not going to show up, or you’re going to show up on some pretty bad terms. You have to get to the point where you have competitive economics to make this work.
Number three, do you actually know what you’re doing in SEM? Google makes it seem like it’s really easy to do SEM, but like everything in life, it’s easy to do but hard to do well.
I’m not saying, “Hire an agency.” You can certainly hire in house, but if you don’t know what you’re doing in SEM, I guarantee you, you will lose a lot of money. Number four, is your conversion funnel as good or better than competitors?
Conversion funnel, in this case, refers to the ad text, the landing page, the post conversion nurture if it’s a free trial or something like that. I guarantee you that, if you test a landing page once a month for 12 months, you’ll see a 20 percent increase in conversion rate.
If you don’t test it, and your competitors do, you’ll be out of luck. Number five. Is your offer as good as your competitors’? Let’s go back to the CRM example. If you’re charging $69 a month for CRM and someone else is charging you $9 a month, the competition’s a click away.
Don’t think that you can trick people into buying something that is relatively the same product for 80 percent more than what your competitors are charging, because it’s easy to find out what your competitors are doing.
With all that being said, if you meet those five criteria, SEM is awesome for SaaS. We have companies that are spending $1 million a month on SEM and SaaS, and are profitable, and overjoyed. Loretta, what’s your perspective from the…?
Loretta: I think that was a really depressing opening. Nothing like I expected.
David: Thank you.
Loretta: No, just kidding.
David: If I spoke slower, would it be less depressing?
Loretta: No. I’m assuming that because you guys are here, you have at least three or four of the things, the requirements that David talked about. The only thing that I would add is, his point on, you really need to know what you’re doing with SEM, whether you have hire an agency or whether you do it in house. It’s something that you can’t set or forget.
You have to have somebody that’s in there every day, that’s looking at the bids, that’s looking at the competitors, that’s trying to be creative with the text, landing pages, etc. It is an investment, not just in money, but it’s an investment in time for it to really pay off.
David: No doubt. Next question. I’ll give this to you, Loretta. How does SEM differ for an SMB SaaS company versus enterprise SaaS company, because SaaS is a pretty murky term?
Loretta: If you are running an SMB company, you can use SEM for the top of the funnel. It’s really easy, or it’s really good, to get people to sign up for a free account or to sign up for a free trial. I’ve done that fairly successfully.
It’s pretty much the opposite if you are running a SaaS enterprise company, because what you really need to do is you need to use SEM to promote a lot of your assets, to promote your white papers, to just collect information from customers and then nurture them.
It can be very effective, but they’re two different strategies and two different tactics, that come in that, in both of those different companies.
David: I would agree. I would say that SEM by its nature is a demand fulfillment channel, which means people are raising their hands, saying, “I’m looking to buy something.” Typically speaking, when you’re looking at an enterprise sale, those hand raisers are coming about through inside sales, through trade shows, through nurture campaigns, from promotions, whatever.
You can make it work in enterprise, but most hand raisers are going to be more like consumers, going to be more on the SMB side. By SMB, I would say if your product is less than, let’s say, $200 or $300 a month, the odds are that you’re going to see a lot more volume from SEM. You’re also going to have the opportunity to convert a lot more people directly in SEM.
I agree with you on the point of enterprise, that if you’re going to use SEM for enterprise, you need to do things that are more mid funnel, which would be things like free demo, free white paper, free trial.
Loretta: E-books, assets.
David: Assets, etc. Maybe I’ll incorporate this into this last question, this one. Do testing text ads matter? Absolutely. One of the things that Google tricks people into is the notion that the higher your click through rate, the better you’re doing. To give you an example, if I had an ad that said, “Free CRM software. Zero down, zero a month, get it for free.”
Click through rate’s going to be through the roof, but the conversion rate in terms of what gets you guys your salaries every week is going to be very bad.
You need test text ads. It’s not just for the click through rate, but also for the conversion rate. Just as I said with the landing pages, if you test your text ads over and over and over again, and I usually recommend you start with A/B testing, where you literally are just trying totally random text ad A, and totally random text ad B.
Finding the winner, and then eventually moving to multivariate after that. You will see improvement.
The one example I was going to give, which the previous question was, let me see what I remember. It was SMB competitors on your keywords. We worked with Exact Target before they were acquired by Salesforce and Constant Contact had an ad that said, “Get started with email marketing for $10 a month.”
We said, “Exact Target’s $2,000 a month. We’re not going to be able to compete on that. Let’s turn that strength into a weakness. We changed the ad to say, “Serious about email marketing? You deserve better than a $10 email solution.”
What happened was, we got all the enterprise customers, and for a while Constant Contact, they actually removed the price from their ads, because we were calling them out as being the cheap solution.
I tried to answer two questions at once there, but what’s your thought?
Loretta: I’m a big fan of testing. I love testing. When you’re trying to create your text ads, I would recommend that you try to put emotion in them, you try to make them very human. That’s what attracts people.
A couple stories. Back in the day, when I worked at Insightly, one of our best ads had a word, “Insightophiles,” in it. You obviously love Insightly. Our CEO hated it. He thought it had bad connotation of the ophile.
He couldn’t really argue with the data, because literally for six months, it was our best performing ad. What happened was, we were working with an agency, and one of the agencies came up with this term of, “Insightly is loved by grandmas to geeks.” Literally, then, that was our best performing ad for two years.
My point is, play with your text ads. You never know where your good ideas are going to come from, and then just test, test, test, test, test.
David: I would say that keywords are zeros and ones. If the word Britney Spears sells CRM software, I’m going to buy it. I’d say the same is true for text ads. When you let your CEO choose your text ads, you get into the HIPPO situation. For those of you who don’t know HIPPO, it stands for highest paid person’s opinion. That’s not a scientific method.
Loretta: The other thing I was going to add on that, in terms of testing, is that now, at Delighted, we do customer experience with NPS. NPS is Net Promoter Score, if you’re not familiar. In our case, in many cases when people are searching for NPS, it’s actually National Park Service. Again, that just goes to, make sure that you test.
Make sure that when people are searching, they’re actually searching for the intent. That’s what it’s about.
David: I can one up you on that story, because I used to work for a furniture website, and we sold bedroom furniture, and one of the things we sold was nightstands. I learned through experience that the word one night stand is not someone looking for a nightstand.
David: You have to be careful of double entendres. Even more seriously, if you’re a cybersecurity company, the words Internet security, 99 percent of the people who typed that in are looking for Norton Antivirus. You do have to be careful about the double entendres, the B2B versus B2C.
I think this is my chance to ask you a question, which is, “No demand channel is an island, so how does SEM fit into all these other marketing channels that people are using?” Facebook, SEO, affiliate, whatever.
Loretta: I think they actually all work together. One of my old CEOs would always come up to me, and he’s always be like, “What’s our best demand channel? What’s our silver bullet?” My response to him is, “Our silver bullet is when all of our demand channels are working as well as they need to work.” They actually complement each other. Nowadays, people are using research online to get information.
For example, you need to make sure that all of your messaging is cohesive across your display ads, across your social, Facebook ads, and then also within your SEM text.
One of the things that I’ve done in the past, for example, is when we bring up a new feature in the product that we want to promote, we make sure that, across all of our campaigns, content, blog, SEM, SEO display, it’s all consistent, so that no matter how someone actually comes to you, whether it’s maybe they see one of your display ads and they don’t click it, but then maybe a week later someone reminds them of something.
They say, “Oh, yeah. Let me search for that.” They’ll see the same message consistently. That’s what I would recommend, if you’re investing in multiple channels, to make sure that they’re all consistent.
David: I would say that SEM can work great, but SEM only works if someone found out about your product somewhere earlier in the funnel. There are three horsemen of the attribution apocalypse. When people really start using attribution for their marketing, the things that are going to lower in value are brand keywords, retargeting, and coupon sites.
Probably not many of you work with coupon sites, but you probably all have brand keywords and you have retargeting, and you think, “Oh, my God! The conversion rate is amazing on this!”
No one wakes up in the morning and says, “Hey, I wonder if there’s a CRM company called Contactually? I’m just going to type that in. Oh, it is! I’m going to buy that product.”
There’s things higher up in the funnel that influence that. I think that, the other thing I’ll say is, with search in particular, I usually say there’s three types of companies in search. There are established brands, there are low price leaders, and there’s the muddy middle. You become an established brand by demand generation, by branding, by trade shows, and speaking, and everything like that.
If you’re in the muddy middle, no one knows who you are, and you’re not the low price leader, and someone runs a search for CRM software, the odds are they’re not going to click on your ad. They’re not going to convert on your ad, because they have no reason to believe that they should trust you.
All that top of funnel stuff ends up impacting your SEM performance.
Loretta: I just wanted to add on that. I’ve worked at a lot of startups, where the brand is not a household name. A couple lessons that I have learned in that particular regard, we used to do, “Oh, we’ve got to do all these landing pages. We’ve got to start with these landing pages.”
And what we’ve found is that taking people from SEM to our home page was the best performing page. What that told us was that people didn’t know our brand, so when they were clicking on it, they didn’t want to go to someplace that said, “Sign up for a white paper,” if they didn’t know what they’re actually doing.
What they want to do is get to know more about the company. What you want to do is not necessarily try to attract, get them to give you their information right away, but make it a little bit of a journey. Literally, for probably two years, our home page was our best performing landing page.
We tested a lot. What happens, we started to see that our best performing landing page was our pricing page, which then showed us that people were becoming more and more familiar with our brand. Our brand was getting out there. Now, people are interested in the price of our product and that sort of thing.
It just goes back to the testing, and if people don’t know your brand, if you’re in the muddy middle, make sure that when they’re clicking on your SEM ad, that you’re taking them on a journey, not immediately asking them to give you something.
David: I’ll add, there’s a great article that came out in Search Engine Land two weeks ago, called “All Keywords Are Brand Keywords,” by Larry Kim from WordStream. I’m not going to explain it in detail. It just shows you how you can measure the impact of brand on your non brand keywords, which is pretty interesting.
Loretta: Our next question. My LTV metric, LTV meaning Lifetime Value metric, takes a long time to figure out. Should I just optimize my SEM spend to a cost per acquisition, to a CPA?
David: CPA is a great place to start, especially if you have a long latency to purchase or you have an LTV that lasts for months on end. The problem with CPA is, to give an example, we were talking before about free software.
If someone types in the words, “free software,” versus “buy software now,” as a keyword, the free software keyword is always going to have a better cost per acquisition, because people are looking for something free.
They’re going to sign up for something that’s free. However, the LTV on anything that has the word free in it is generally not so good. You absolutely need to use CPA as a metric, but if possible, you want to blend that metric with more downstream behavior. One example would be if you have marketing automation, and you have some lead scoring.
Calculate the lead score at a keyword level. Another thing that’s really important is to take the keyword that got a conversion, but also pass that keyword variable into your sales, into your CRM, and match the keyword with the revenue that that keyword actually drives.
A lot of people don’t do this. They don’t pass a variable into their CRM to look at the LTV at a keyword level. If you’re doing it at a channel level, just AdWords for example, you’re not going to see the variances in how those CPAs turned into actual LTV. It’s not an out of the box thing to do, but there are companies who can help you with that, and any expert can help you with that, as well.
Loretta: Definitely. In marketing, you’re spending money. At the end of the day, you have to be responsible for the return on that money that you’re spending. It’s very important in all the companies that I’ve worked in, is what is the return on that spend?
It’s really important that you know all of your metrics, and I’ll just go off in tech jargon. It’s really important that you know your customer acquisition costs.
That, you know your CAC ratio, you know your Lifeyime Value. You need to know, really, throughout the funnel, what that demand channel does. I would recommend that you try to get those metrics as quickly as possible.
If you can’t, you can use a proxy, but you really need to know what, you’re putting this much money in. How much money are you getting out of this investment, from a marketing perspective?
David: I’ll just add, I refer to this sometimes as information asymmetry. If you know that the word free software has a $500 LTV, and the word software has a $1,000 LTV, you end up bidding double on software.
Your competitor who doesn’t know this, doesn’t bid double, but bids whatever they bid on free software, what ends up happening is they outbid you for free software because you don’t want it, and you outbid them for software, and suddenly they can’t figure out how you’re making money at this.
You have better information.
Loretta: That’s true. The other thing that you could even take it one step further, which is what I’ve done in the past, is if you have the type of analytics programs that actually let you know what people from SEM are actually doing in your product, what you can do is, you can take that information and then turn it into more text ads.
You’re saying, basically, people from SEM do this in the product. If there’s any commonality, then take it, and take it again to the top of the funnel, and to try to attract more, so it’s a life cycle thing.
David: Absolutely. Virtuous cycle.
Loretta: Virtuous cycle, there you go.
David: Am I asking you this question?
David: I have a limited budget. If you have a limited budget, should you bother with competitor terms, or even non brand terms?
Loretta: I just have a philosophy that if you have a very limited budget, that you probably shouldn’t bother with competitor brand terms, because to me, frankly, I just think it’s a pissing match.
If you have all the money in the world, that’s a completely different story, but I’ve been in situations where Salesforce, for example, has been bidding on my term, and they have about 70 percent impression share.
In that case, I would say, yes. I would bid against their term when you can, if you have the money. For the most part, it’s a philosophy thing. There are plenty of people out here that might disagree with me. For brand, my philosophy is generally, unless I have leftover budget, I would not bid on other brands.
Non brand, for sure, no question. I would definitely bid on non brand terms, but that’s just, like I said, my philosophy based on my experience. Typically, I don’t like to do it unless there’s an aggressive competitor that’s bidding against me, and then I will do it. As a matter of course, for brand terms, I typically don’t.
David: I would say, if you are the industry leader, the 800 pound gorilla, you should definitely try to get some friendly fire agreement with all of your competitors, not to buy each other’s keywords. If you are a newbie to the space, you should try buying your competitor’s keywords, because they’re going to get a lot of traffic.
Google is going to penalize you by reducing your quality score, and even if you’re the only person advertising on that keyword, you’re going to probably pay $10 a click, but we call that conquesting, and it can be a really effective way to drive revenue. If you want a little advanced tip, if you haven’t tried Gmail Sponsored Promotions, GSP, that’s a way you can actually advertise against someone’s email list.
That’s effective. Non brand, I think, is a no brainer. As I said before, I think brand is the most overrated thing out there. If your brand is always going to get more credit than it deserves, non brand’s where you really win the day. I think we have time for one more. Last question. We’re more or less on time.
David: Everyone knows about SEM. There’s no arbitrage opportunity anymore. Can it really drive massive ROI?
Loretta: Yes. Last year, the company that I worked with, basically our SEM channel had a customer acquisition ratio, a CAC ratio of 3.3, and we saw about a 275 percent ROI. Granted, it’s not 1,000 percent, etc., but it was pretty good. We were very happy with that.
To give you context with those numbers, in terms of the CAC ratio and the 275, it was probably in the top third of our demand channels that we had. We did content, SEO. We did display. We did a lot of different channels. For us, it was a worthwhile investment, but again it’s something that you have to…It’s not something that just happens overnight. It’s something that you have to nurture.
You have to pay attention to actually start seeing an ROI on it.
David: I think the arbitrage opportunity is gone. It’s a lot more competitive, but there’s a reason that SaaS companies spend half a million, $1 million a month in SEM, and that is because when you measure properly, when you do the right testing, when you have the right experts, it is massively profitable, and it’s an incredible way to scale your business, once you have the right pieces in place.
We’re out of time. Loretta and I will probably be wandering around out there if you guys want to talk to us. Thank you very much for letting us talk fast.
Loretta: Thank you for coming, as well.