Hey everyone, welcome to Stacking Riff. I'm Sam Kean Lee and today we'll be chatting with Ashley Lulens about paid marketing's influence on organic marketing results. So many marketers and business leaders look at paid media in a cyber today. If we put an X dollars in the paid this channel, we should expect to get Y out of that paid channel.
What if we said that there's a really strong correlation between paid media and your organic and direct results? Would you believe us? So every now and then marketers get that gut feeling because there's a connection between things but because they can't prove them, because they don't have the data, they fall by the wayside. Ashley and I pulled on this thread to see if there's a connection that exists and what we found might just surprise you.
So Ashley, you recently shared an example on LinkedIn of how paid efforts impacted organic and direct results. Around the same time, another member of our team surfaced the same type of insight to me regarding a different client. So do you think there's something really that can connect these two? Yeah, I absolutely do.
And I think the really cool part about this is that Sam and I will be digging into two very different use cases with data. So support the sense that you know we have that gut feeling but we have the data from two very different cases that we'll be digging into. So I definitely think that there's a correlation. I love it.
And what's the use case that you're going to explain today? Yeah, so my use case is very rare. I will just call it off right away. Not many companies will shut off paid for a period of time.
And this is the case that I ran into where paid completely sought for a couple of months. So you really got to see the correlation there. There was no ifs ands, what's about it. So got to see it from that point of view.
So just give you a little bit of background. So all paid efforts were paused for roughly three months. So in that time frame, I pulled a little bit of data to support this hunt because I was watching you know Salesforce and all the great stuff that we're constantly watching for. And also I was looking at a little bit of site traffic as well because I really wanted to see if this hypothesis was true.
So in this time frame, I looked at organic and direct lead sources and Salesforce for deal created and that fell for roughly 50%. So just from shutting off paid, organic and direct lead sources for deals created fell by roughly 50%. That is an astounding number I feel like. And you can really see that strong impact there.
This was taken the 50% was from previous month averages. I'll just say that too. So I was looking at like what were the previous months that I was comparing it again versus that time frame and that was where the average came from. And then I also looked at organic and direct site traffic and saw that that decreased by 40 to 60% as well from the previous month averages.
I found really, really interesting as well because we're talking about deals created, which that can have a lagging effect of course, but we're looking also just at site traffic. And so I think this really also really leans into this critical need to look at marketing in a blended view instead of looking at it in a siloed channel lens, which I think we're all somewhat guilty of. It's kind of where marketers are really faced right now. It's like, cool.
Does it really impact organic? Of course, it's the direct lead sources go down, but is it impacting? And this is a really great case to say, hey, you really need to look at it at a blended view instead of in that siloed channel lens. Marketing is a multi-channel machine that works really well together to get to your ultimate goal, which is of course customers, our favorite metric.
And if you only look at that last touch attribution lens, you're going to miss out on a lot of growth and dismiss being customer focused, too. You're only really looking at that. You're not thinking of the customer. You're not thinking of the customer.
You're not thinking about how they want to engage with content and education in your brand as well. And so I would also say that this effect that I saw when shutting off paid into, you know, site traffic and deals and all the great metrics that we're looking at a lot, I would say this also really reflects to how we look and see success here at Refine Labs with clients when implementing paid strategies. We typically see an increase in organic direct and branded page search results in correlation of when we start our Refine Labs Playbook as we like to coin it with the paid strategies. And it's usually a little bit of a lag period, too.
So if we implement it, we usually see a lag period. It really depends on the average sales cycle, but I'm going to just say roughly about three ish months if we're looking at a 30 to 60 day sales cycle. So I think this also is a really strong correlation with what we see typically with clients. It was cool to see it with this hypothesis of turning off paid and all that great stuff.
I would say also, like breaking out by channel is great when you're trying to analyze trends for insights, but it shouldn't be the main inputs or reports for goal setting. You really have to look at this holistically and from a blended approach to see the really, you know, the strong effect spends has on organic. I'll keep going a little bit before I do that. I'll take a breath and Sam, do you have any questions on those results that I just kind of rambled off a little bit there?
Yeah, I mean, you ran the test that I've always wanted to run, but I've never wanted to run. And it's like, well, let's shut it off. You think that marketing is improving and if you're doing anything, I'll turn it off. Like, you never want to do that.
But what's also so funny is I always remember looking at former company, we implemented a multi-touch attribution and we started to break out and see like direct organic. Oh, it carries so much weight, so much play plan, so much revenue. And they think, well, how do we invest more in direct and organic? We need to do more SEO.
When you explain this, I think, does the reverse of that conversation ever happen? Where is SEO not working anymore? Do they immediately go to thinking like those channels aren't working? Well, the reality is one big variable changed.
We didn't change our website. We didn't change any of the content that was producing. It was just, are we showing up in front of the right people at the same rate that we used to be before to create that awareness? Yeah.
No, I think that's so important to notice. And it makes me reflect back on early in my marketing career too. I remember evaluating data with a leader. And we noticed, of course, direct and organic produced the best type of leads for the company.
It was the most profitable, all the great things. And I'm like, cool, we can't really drive that though. That's not a lever we can necessarily pull. And if I could go back to early marketing career, Ashley, I would say yes, you can.
It's just not going to be in the way that you were taught. You know, with the 2008 marketing playbook, it's a little bit of a more of a correlation effect. And you have to look at it a more high view. And so if I could go back, I could say, yes, of course you can impact that direct and organic.
It's just going to be in a different way. And it's running your marketing in a way that centers truly around the customer. It's running paid. That's not trying to get someone to fill out a form, either natively on that paid platform or on your site with an e-book.
But honestly, nobody reads. I remember even call. I've done SDR calls on leads that I had previously acquired. And if you really want to test it out, I strongly encourage calling leads that you are acquiring because I'd be like, who is this company?
What do you do? I downloaded what? So yes, you can definitely impact it. It's just going to be in a different way.
Yeah. Oh, man. Giving me flashbacks to my SDR days. The hardest job I've ever had right there.
I'll tell you that. So yeah, as you talk then about how you tell, you know, previous Ashley about that, it sounds like you would take that insight and almost say your attribution isn't quite what you think it is relying on that. So it sounds like you might have a thought around what you do if you could go back and do that now. I feel like we could spend this until a little bit of a teaser to go and listen to our content on hybrid attribution because I think that is a critical piece in the puzzle here, taking the qualitative and the quantitative data together.
So you're asking your customers, how did you hear about us? So you know, we run this test here at Refine Labs and we get a mass majority coming from organic or SEO. We don't do organic and SEO here. And so when we also compare it against the how did you hear about a field, it's these page channels that you would normally write off or that we stopped doing, you know, it's podcast, it's LinkedIn, whatever you have it.
So I would definitely recommend doing that. The other piece is which I know will cover at the very end and I want to make sure it's all packaged together. You have your notebook and pen out, you can put it all together, but you really need to look at your metrics blended. So what that means is holistically.
So you're going to take all of your marketing spend and you're going to divide that out and figure out your marketing, your marketing, your marketing, your back period, your blended cost per opportunity created. All those great metrics and you're really going to look at it as a high level view because at the end of the day, it doesn't really quite matter how someone gets to your funnel. It just matters if they become a customer. So I really highly encourage everyone to look at marketing in a blended view and a holistic view and then to also evaluate doing hybrid attribution because it will, you know, help answer a lot of those attribution questions that we've been trying to answer and that we got a little bit of insight on with shutting off pain.
I will say, so since we did shut off pain, we have turned it back on recently. Nothing quite to update here. I wish I had like a impactful number to tell you all, but nothing quite yet. I will also know it's about a 90% decrease on the spend that we were doing.
So you know, it's not going to see the same impact which was going to lead really nicely into Sam's experience. One last thing that I'll call out because I think it's really important, especially when you are talking about completely shutting off pain, is, you know, marketing the investments or the lack thereof that you're making in this exact moment today are for the future. Most companies won't see a direct hit immediately when paid stops. However, they're really going to fill that pain in a couple of months and that was true in this case as well.
You know, on day one when paid stopped, it wasn't like the engine stopped. However, just, you know, it's like you're running out of gas on the highway in your car, it just slowly starts slowing down until you're at a complete stop. So I would say that's also something important to remember, especially as marketers when you're talking to leaders and so on, is the investments that you're making today are really for the future. With that though, in talking about like, you know, stealing the spend too and how that has an impact on direct and organic, Sam, I know you have a really cool scenario as well.
You saw a little bit of the flip side where it wasn't that you shut off paid completely, but you know, saw different percentages of the spend and how it really impacted the organic and direct as well. So I want to hear your case and what you saw as well. Yeah, so this one was interesting in the sense of we look, about a six month window, we were fluctuating in terms of the spend. Someone's we'd increase the spend, someone's we had to decrease the spend.
So all recently, so the same reasons that everyone's been seeing today, you know, economic conditions, we want to increase our pipeline quick for our growing sales team. So for many of us who are working with a sales experience, when you think like, get a demo and talk to sales, we often feel like we see a correlation between increase and decrease spend and the business results. But when you're in a space where it's something like you could be spending $1000, $3000 for demo, if you don't have a large budget, it's really hard to tell what turning that budget up or down by 10 to $15,000 per month means because you're looking at, okay, maybe a difference of one to six demos. Is that significant?
It's really hard to say, and there's often other variables at play. So this is where looking at data that's either a product led, so a free trial or a dual motion, so sales led and free trials do market can help since the barrier to the hand-reaser is a lot lower. So in this example, I'll talk about data from a dual motion, go to market where they offer both the get a demo experience as well as a free trial motion. And so to lay out the data, what we did when we were analyzing this, we looked at some core metrics, it was really just three big ones that we'll focus on.
Add spend, the reach, so to set campaigns as objective, effectively it becomes a factor of spend, and then the ICP hand-reaser, so the sum of the people who either requested a demo or signed up for the free trial. Because they had the dual motion, we could see very clear correlations from changes in spend and reach levels, and we used a 30-day window post change. So it took us a few months to start to see this, but it was because we had these fluctuations in drastically increasing the spend and then having to pull back that we saw the post 30-day kind of the wind down the highway as you said a little bit ago. So what we looked at was the baseline before us.
We found that they were spending about enough to reach 83,000 people per month and it would generate 49 hand-raisers. So I like to use that just to level set on, okay, previous spend and then as we start to go up and down, what's that mean? So in month one, what we did was we increased our spend to get in front of 132,000 people. So this is a 59% increase, and that drove 104 hand-raisers, so over 112% increase.
In this time, we also created new audiences which were really focused on tightening the ICP targeting and wanted to evaluate that moving forward. So the combination here was spend plus more effective targeting. Month two, what we wanted to, not so much what we wanted to do, but we were starting to feel the beginning of this economic impact. So they said we didn't quite have the funds to continue the same pace.
So we had to pull back spend a little bit, it was just about 8% decrease and we saw over the next 30 days that that led to an 11% decrease in hand-raisers at that time. So coincidence, I don't know we're wondering what's going on here at that time. So month three really had to pull back a bit on the budget, pull back another 23% and then again we saw the drop in the hand-raisers. So clearly this connection going on at this time.
So we shared this with them. We said, hey, we've got this theory. Your spending reach are very closely correlated or appear to be with the hand-raisers. Are you all open to testing it?
So they go to their board or not the board, their executive team and asked for a little bit more budget. So we basically went back up to the same amount of spend that we were in month one. So increased the spend 37% from the previous month and over the next 30 days we increased hand-raisers back to 113, so 40% increase. So I mean, it's crazy to get this type of immediate feedback.
We're like, okay, we're on to something. Month five, 15 pushing it. Is this going to fail? So we increased our spend even more, 13%.
We saw another increase in hand-raisers by about 10%. So we got to the point where we could almost, with confidence, say for every 1200 people that we reach before diminishing returns. So you can only reach so many people in LinkedIn for your audience size, but we knew that we could drive a new ICP hand-raiser from that. So it's a really cool experiment.
I think it's a little bit easier because the barrier to entry was a little bit lower with the dual motion, but to the point earlier, like, does spend influence overall impact because these are not direct response off of social. We were seeing people come in more often than not direct, organic, all the other traditional channels, but they did implement the hybrid attribution we were talking about earlier. So we did start to see LinkedIn show out more and more as we had the months where we were spending more. So it's a crazy example, but it's really nice and we can validate some of these things that for so long we wish we could tell our previous selves.
There is something there you just have to pull that through a little bit more and know what to look for in the data. That's such a cool experience and experiment. I'm a little jealous that you got to be a part of that because, like you said, it's always something that we're looking for, is to see that strong correlation and to be able to report back on it in that strong of a sense as well. I'm really curious what the hybrid attribution is too.
We had this hypothesis. We were able to ask for a little bit more budget from executives as well. How did you use that in proving your theory as well? Was it part of it?
How did you use that in the story around it and how did you use that for your hypothesis? Yeah, so the hybrid attribution came a little bit later after we made that ask that was used more to justify what we were doing. So it was an additional flag in the ground to say, this is working and now we can validate it not only from your correlation, but we now have your prospects coming in and we're seeing at a higher rate, they're talking about LinkedIn, Facebook, these other places where they're learning about you, but they previously weren't talking about you. We're getting the traditional online research or my colleague mentioned you something like that, but there was definitely that supporting trends.
Now when you've got two different data points validating this, it makes it harder for them to ignore when you want to go back for whether it's an upcoming budget year or something else to say. If you pull back, you know what's going to happen. I love it. I'm also curious, how are you using these findings for future budget planning and goal setting?
I think that's such a cool correlation. I think a lot of operators and leaders will be able to cling to that as ways to pull different growth levers within their plans of goal setting for the next year and things like that. I'm really curious how you're using it as a kind of like a go forward plan. Yeah, so this would be something like that final insight where we say, we've got your ICP diagram and we know that for every 1200 people that we reach in a month, you're going to get a hand raise around a bit.
So say their audience is 100,000 people, we know it 30 to 40% will be on LinkedIn at a time frame. So if you really want to grow, that's see how much we need to spend to reach 30 to 40% of these people each month and then you can work backwards from okay, if that reaches 40,000 people, takes 1200, that means we're going to get whatever mental math comes out, just under 30-ish hand raises. So not the real numbers in this case, but that's how you can start to back into something like that because one thing that marketers sometimes like to pretend doesn't exist is a point of diminishing returns. Spoiler point of diminishing turns does exist.
So you can't quite ignore it. You can use that to think about, okay, if you can get in front of this many people per month and continue to really build a momentum, that number is going to get more efficient over time as well as more people who are a member of you because the other thing is they are in current contracts, their budget seasons aren't always hitting at the same time. So you're going to get more efficient the more that you stay in front of them and have it become less of a U versus a competition when they end up in the market trying to figure out, I have a problem, I don't know who to go to, now I have a problem and I know who exactly to go to. So 1200 numbers will become 1100, become 1,000 and should hopefully continue to really pay dividends over time.
Yeah, and I'm really glad that you talked about the point of diminishing return too and you were able to run this test to really test that theory out too because there is a point of diminishing return and there's a point where you will cap on a certain channel. Typically, we'll have your MVP channel, so say it's paid LinkedIn and then it's, you know, let's say Facebook or whatever, maybe really running these tests of kind of the throttling of the spend and seeing the correlation and then figuring out where your point of diminishing return is. I think that's a really interesting experiment to be running and to also prove out for your company is to figure out where you cap out in a channel and then when you need to look for your next step of growth. So which, when is it time to get to your next channel?
When do you need to implement something like Reddit or Twitter or whatever? Maybe that's a whole other conversation that we won't dive into because I think we would spend hours talking about that. Yeah, exactly. And those are just the pay phones that you mentioned.
So everyone's websites really optimized as well. So there's probably so many different things that we can be doing with their conversion points and everything else to even make that working better. So for those listening, I think let's talk a little bit more about like how can they spot it in their own data or how can we make this actionable for them. So one thing I was like to look at is attribution versus correlation.
So attribution, as we spoke about earlier, it's helpful, but minimal data is really going to be passed through. Sometimes you'll start to see a few user conversions if you apply you through and a platform like LinkedIn alongside the click through conversions. It's helpful to get insight into some of the audiences, the campaigns that you're running, the content types, but where you'll start to see it become more useful for you is the correlation side of that gut feeling that you sometimes get, but you just quite can't prove. And so this is where you need to start rattle pulling and see if there's something there, which is what we started to do here.
So in the example that I shared, one of the things that I was looking at was like AdReach and its correlation with direct and organic traffic. So alongside HandRazors, we did see the overall traffic was coming up from those along with branded organic and paid search phrases for the company's name. AdReach and HandRazors are exactly what we went through. And other correlations that you can look at are things like employee organic sharing and inbound business requests.
So I know there's a number of companies that have maybe one to two people who are very vocal, others that have a number of people on the platform and when you aren't running paid or there's other ways, I mean, that is a phenomenal way to really establish awareness in the market and let people know what it is that you can help them with. So along with those, I mean, if you are thinking about what's working, if you want to understand what's in marketing, what's going on, but you don't want to get into the channel level, attribution in the channel level correlations, how are you approaching that? Yeah, so when I look at metrics, the ones that I'm really looking at is I'm looking at marketing blended. So I'm looking at your total offset you're creating, your total qualified opportunity, so anything with a deal stage that closes at 25% or greater consistently.
And then I'm looking at what the win rate is, all the sales efficiency metrics, so average sales cycle, ACV, your pipeline velocity, all those great things. I'm looking at that at a blended level. And so then when I find something that's either going up or down, that's when I like to ask a further question of why are off sounds. So in this case, for instance, in my scenario, offs went down.
So my first question is I have a hypothesis that it was due to paid. And so that's when I structured the data rabbit hole, figuring out if it was correlated to that. But if it's not that, there's other things that could be going on in your marketing too. It's just really first asking what the hypothesis is.
I think that is critical before you go into a data dig to figure out what's going on is you need to ask a data hypothesis before, otherwise you will spend hours in spreadsheets and your CRM looking for any noise instead of signals within your data. So you can also look at in your marketing, is there anything else that has changed as well? So have you refined your targeting? Have you implemented a new messaging?
Have you updated your website or done any CRR adjustments? Watch the podcast? Any subject matter experts starting to share on social, new fundraising announcement, new product or feature announcements? If you go on and on, you also need to look at what other variables are at play to figure out what it is.
But I think it really truly starts at one having your core metrics. You need to make sure you have your strong KPIs that you're marketing is being judged against. And if it's leads, that's another conversation. Take a pause and look at the pipeline metrics there.
But look at those. And when you start to see something go up or down, I encourage people to form a hypothesis and then start to go down those and figure out why. And so I have a hunch, which by the way, as a marketer, when you have hunches, those are very strong. And you should really lean into those hunches because sometimes it's the art of marketing.
You're taking the science and you have the hunch, which is the art. And go down those rabbit holes to figure out what it is. So to peel back the layers a little bit. So overall, I would look at the blended.
Also if you're are throttling paid or if you're shutting paid off or if you're not sure if paid is working well for you and your company, another thing I would look at is efficiency metrics. So what's your ad-cack ad-cack payback period, things like that? If your ad-cack payback period is over six months, then that would be assigned to me that something is not resonating within your paid or there's something else going on too. So then that would form a hypothesis of, hey, I don't think our targeting is working or hey, I don't think this is the right channel, whatever it may be.
So I also look at those efficiency metrics. Which metrics am I missing that you typically like to look at as well, Sam? I think you've covered the main ones. And this is a question that I get asked all the time.
When you are looking at these different metrics, I know we said, you know, lead metrics, different conversation for another time. But when you're starting to look at pipelines down, payback periods, are you looking at those weekly, monthly, quarterly, what's your time frame to say like, what's the appropriate time to before you should be reacting to it or starting to form a hypothesis? Yeah, I actually think this really depends on the company and how fast your pipeline moves through. But typically I like to have yearly benchmarks.
So I like to use the company's data to have yearly benchmarks that keep in the back of my mind or in a spreadsheet open to compare against. And then I like to look at it quarterly. And then I like to also have a pulse on it, monthly and by monthly as well. That doesn't mean that anything needs to change.
Business is too volatile to make weekly changes. If you're going to look at it a weekly basis, you're going to hurt yourself with growth. If you are constantly like, oh my gosh, we're down. We need to fix this.
It could just be a myriad of reasons of why it's down in a week. I would not panic. Hit the panic button just quite yet. I would wait until you see the quarterly.
But also look at it monthly too. I do think that's important. Like if you notice a very strong trend that's happening on week over a week into your monthly reviews, then I do think that's a time to not wait until it's too late and to make a call. But I do think it takes a lot of judgment, strong judgment in order to make that.
But typically I recommend look at your data and a yearly benchmark and then really evaluate it quarterly as well. Yeah, I like that because that's something we get trapped into is being able to recognize is it a blip that you bounce back from or is it a trend line that you need to start to course correct. So cool. I think this was probably one of my favorite conversations today, getting into the data and really nerding out over all this.
So I guess to sum up, some more conversation today, it's one paid absolutely influences organic performance. There's got to know how many different variables that can impact the relationship. But as you said, get curious. Start asking questions, form the hypothesis and then see what you can find.
But as you said, those functions are usually on something. So this is not actually, I'm glad I got to catch up about this. Yeah, I had a blast too. I feel like this is every marketers entry and conversation here.
I got a lot of feedback when I talked about this on LinkedIn. I think I got, oh my gosh, I'm not alone type of response. I've had this idea, but I've never been able to have the data to prove it. So Sam and I did the work for you.
You're not alone. There's definitely a correlation. Love it. Well, this was fun.
All of us we continue to come across more of these examples, but great catching up. Thanks, Ashley. Thanks, everyone.