Making Sure Your CAC Doesn't Kill You: Driving Low Cost Customer Acquisition

So this is our last session, I believe, of the tactical stage.

Super excited about it.

The title is Make Sure CAC Doesn't Kill You: Driving Low-Cost Customer Acquisition.

Super excited personally to hear this one. So the moderator today, I wanna invite her up to the stage, is Joan Chen, a partner at Foundation Capital. And if you wanna go ahead and bring up the panel as well. Let's welcome her to the stage.

Thanks everybody for coming. I'm Joanne. I'm a partner at Foundation. We've been investing in SaaS for a very long time, so I'm super excited to be here. This topic, the cost of customer acquisition, is also a favorite of mine. I work with 5 SaaS companies currently, and I know that we discuss this ad nauseam. In fact, I came from a board meeting this morning where we spent probably 30% of the time trying to brainstorm on how to lower our cost. Now why is this such an important and interesting topic? I think this topic was largely ignored in 2014 and 2015, and we saw some of the consequences of that. Startups have the mentality of grow at all costs, putting a lot of money into growing top line no matter what. And the result of that, CB Insights actually started a tracker tracking how many companies raised down rounds since 2015, because growth and revenue is not all created equal. Luckily, today we have 3 amazing panelists who are experts in thinking about lowering cost of customer acquisition, and we're now smarter, and hopefully smarter because of these guys. So I'll let each one of them introduce themselves. They all come from very different companies and different perspectives, and hopefully you guys will find this valuable. So maybe Guillaume, you can start.

Hey, so I'm Guillaume. I'm the VP of Growth at Segment. So Segment is a platform to help you collect your data from where it happens and route it to where it's valuable. Prior to that, I worked in various startups as head of growth, and I started my career at Apple in online B2B sales. 12 years ago now.

My name is Jeff Samuels. I'm glad to be here. I currently run security marketing for Cisco, and I came to Cisco via the OpenDNS acquisition. And at OpenDNS, I was chief marketing officer and really thought about this topic and concept a lot there. And yeah, I'm looking forward to having this conversation with these great panelists.

Hi, I'm Sunil. I run marketing at Replicon. We're an HR software company. We sell time and attendance solutions and professional services automation Solutions. Emergence is one of our investors. Prior to Replicon, I was actually an EIR at Foundation, and before that, I was co-founder and CEO of a company called Scripted, which is a marketplace for businesses to hire freelance writers. And by night, I'm a— I write satire. I have a digital publication called The Bold Italic, which some San Franciscans might be familiar with, and I wrote a series called This Is Your Life in Silicon Valley, which which was fun.

Sunil's being very modest. I think his satire pieces have reached a million people. At least one of the pieces did really well. So maybe to start, how do you guys think about defining low cost of customer acquisition? I know it's very different across different organizations. But Sunil, maybe you can start with sharing with us how you think about it.

Yeah, I mean, like all of you, I mean, we use traditional metrics, CPL, conversion, et cetera, et cetera, et cetera. You know, everybody is in search for that mythical channel that's going to drive down acquisition cost. But my big thing now is, you know, what if we spent all of that time that we were searching for that mythical channel and instead focused on the tactics that actually work? And so trying to drive down acquisition in very traditional ways rather than searching for something that doesn't exist necessarily.

Yeah, I look at acquisition, customer acquisition, a little more holistically. I look at it like it's not a marketing metric. I mean, it's great that there's so much focus on it now and it can draw a crowd like this, which is really important, but it's the way a company breathes. And if you look at, if you start, especially if you're an early company and really starting to think about the technology, I always encourage startups and even larger companies like Cisco to start thinking about go-to-market and customer acquisition in the product development stage. Again, everyone should have role model companies. You think of an Atlassian, a Splunk, a SolarWinds, where amazing on customer acquisition. I mean, the metrics on Atlassian are off the charts, and there's been a lot written about it, but it's about reducing friction of the customer coming on board and using the product. I mean, that is where it starts. Then you can amplify that via great tactics like Sunil was saying.

Guillaume, what do you think?

Yeah, so I'd love to disagree, but not yet. 'Cause my main focus is not usually on finding the cheapest channel, nor on finding the biggest channel, but on actually fixing leaky buckets. Most companies that I work for, that I consult with, have very leaky buckets, And you're just pouring more leads into a website that does not convert. And I really love actually finding a way to patch that bucket and have a very high conversion rate. And then any cost flies if you convert a very high number. So I actually like to take the opposite stance and say, I'm ready to pay more, actually more than you used to, because I can score with some tools. And we can talk about that. And then I can convert convert those people better. I like to look for the best customers than the cheapest customers.

Absolutely. Leads are not created equal either. And I think the leaky bucket problem is something that we've all faced at some point. So thinking about how do you measure success in this realm, how do you guys think about success? What are some of the structures or tools or methodologies? And how do you communicate that to your companies?

Yeah, so for me, success is beyond just the cost of acquisition or the return on investment. Success is the great customer experience. We're trying to pull from the market and pull from all of those channels, those potential customers, and we're creating a friction, as you said, or a cognitive load. And so if you can make that experience as good or as fantastic as possible and actually reduce the friction, then you're gonna have by effect a great conversion and by effect actually a great return on investment. So that's my metric of success actually. That's what I optimize for.

Any different opinions on what is the most important thing to look at?

Yeah, I mean, I think it's, if you look at the title of the panel, like that is a company-wide metric. And if that's what you're focusing on, I think what she's saying is part of that, it's sort of too late. You have to look at the elements that go into that. On the website, you have so many different journeys that you are paying for, that you're driving organically, and you need to look at the micro journeys and the analytics on every single one of those to know where to invest more and where not to. All of those things drive the larger top line metrics that you're gonna show your board. Again, you wanna be looking at these kind of data. I'm like, okay, looking at SEM and looking at the amount you're spending for search and how much revenue you're actually getting, connecting those dots and connecting those lines together. And then for all of your journeys, 'cause each one's different, each one's gonna have a different content journey along the way, and there's gonna be different investments to optimize.

[Speaker] Yeah, I mean, look, we look at all the traditional metrics as you would expect by channel, et cetera, et cetera, et cetera. But at the end of the day, as a company, if you have customers that are willing to talk about your product and what an amazing experience they've had with it, makes everybody's job at the company easier, period. And if you don't have that, I mean, that's gonna make marketing's job more difficult, sales job more difficult, just everybody, right? And morale-wise. So trying to focus more on that, and I don't mean just NPS, right? It's NPS plus, right? Customer willing to do webinar with you, spend a lot of time. At the end of the day, happy customers just makes everybody's job easier.

If I flip that question around, are you guys for a minute, what do you think is not the right way to measure success or even just to measure cost of customer acquisition?

Volume, that's the worst. I must have gotten what, like 50, 100 BDR emails from some of your companies. Sorry, I won't tell you which ones stunk. I can't get a laugh out of— I'm sorry, there we go. But volume is not, effort should not be confused with results. And I think there's a lot of, effort-based measurements out there that are not really valid. And volume is not the answer.

Yeah, I won't even try to make a laugh. If someone who writes satire can't, I won't try. The focus on short-term results also can lead you down the wrong direction. It's elements of volume, elements of how can I get more into my database, all these sort of things. And there's the short-term effect of, OK, what that's going to do. Yes, it works. or to a certain degree in the short term, having all these SDRs, BDRs calling and emailing and stuff like that, you're gonna get some in the net, but there's a cost to that. And the long-term cost of that is impacting your brand, impacting the way the rest of the world views it, and the way you want the rest of the world to actually leverage your technology. So I think, again, really focusing on that top of the funnel. There's ways of doing that, but, It is not the end-all be-all.

Absolutely. Guillaume?

Yeah, yeah. I mean, I totally agree, and I can even give like a quick story. Like at Segment, when I arrived, the sales team was doing some pretty aggressive outbound, and they were actually doing what you said, which is burning through the market. And the cost that you talked about is, there's a brand cost, of course. There's also a cost of once you talk to a potential customer with like a pretty bad written email, like something not very engaging, That lead, that potential is lost for maybe a year or two, at least in B2B. You're not going to go back with another email within 3 months saying, "Hey, it's me again." That's not working. So you're burning through the market. And if you burn through all the nice and good leads in a few months, then you're not eating afterwards.

Yeah, another, I think, misconception people have about— I mean, they don't think about it in terms of what it takes to actually develop CAC is that it's marketing and sales. right? And as marketers, you're focused on what you can control, but actually looking at the journey about how your prospect's gonna turn into a customer, sales is an important element of that. Whether, I mean, obviously the website's a big part of your sales process, but in most companies, you're gonna have someone catching that lead and closing it, and making sure you're optimizing that as part of your marketing and your strategy is really important.

Yeah, and try to take the customer empathy thing really seriously. If they're, think about your ARR, And what your average sale price is, if you're selling a $25,000 product to a customer, they're going to take that as seriously as they would buying, I don't know, a Toyota Sienna. And so put yourself in their shoes. And what do they want to hear? And I think that's an important element that just can't be overlooked.

A very valid point. You guys will be surprised as venture investors how many pitches we see in a given year. I probably see around 1,000 myself. That ignore this piece of it. Oh, the customers will come, we build a great product. We don't understand except that there's a lot of people coming to our website, but having the clarity around the top of the funnel conversion, the customer experience, all these things, even just having perspective and hypotheses around that, I think will help you once you go through the fundraising process. So maybe going to the brand question. How do you guys think about the coexistence of brand versus efficiency? Are those two things that you take into account in your organization? How do you think about that in general?

[Speaker:JASON M. LEMKIN] So they go hand in hand. And again, it goes back to the holistic view of the way I personally look at it. I've always looked at the website itself as a product. I mean, it is an application that has user stories, that actually has a purpose, and that's optimizing, and you're constantly developing for it. And if you think about it that way, as a important as the product that you're marketing, that is essential. And if you go too far away, again, it goes back to the short-term, long-term, it will impact the brand. The idea is like, what kind of brand do you want? When you're at a trade show, do you wanna just be scanning everyone that's never gonna convert to Neil's point about the top of the funnel has limitations? Or do you wanna be more focused on who is your customer and making sure you're connecting with them? And there's many examples of companies that, that go too far that way and probably don't succeed partly because of that.

Yeah, the company's brand is no longer the CEO's brand. I think there was a period of time where you would just identify, okay, well, Steve Jobs is Apple or whatever, right? But the company's brand is now the company's brand. I mean, information is available publicly about everyone at your company. People are looking up your BDRs' profiles on LinkedIn. People are looking at your Glassdoor reviews. Is this the type of company that I want to be in business with? So I think thinking of the brand as every interaction with a potential customer or an existing customer is really critically important. And that's just something that's a takeaway on brand that—

And Guillaume, Segment has had this great, great growth recently and also significantly growth in brand equity. How do you guys think about that?

I don't think there's— such a thing as one brand as we move into the future. I mean, I started at Apple, and Apple has like a very unique brand that's able to pull customers to the product, and the customers just adopt the message. But the truth is that most of our companies are not Apple, right? And we don't have that gravity. And so we try to find the best average message, and I think that's a failure on our part because we're not as good as Apple. And the customer comes there and has that cognitive load and tries to understand, hey, does that message match with what I want? And that's a failure because then we have the high bounce rates and people don't understand and they don't convert. So I think that as we move into the future, what we're gonna see is that the brand part of the marketing is gonna come up with like multiple messages. And I'm just not just talking about A/B testing, I'm just, I'm talking like hundreds of different messages for each subset, maybe even each individual, to really customize and personalize the experience as we do in emails right now, so that it really flows without friction even on the website. That's, I think, is the next challenge that we're working on. And at Segment, that's what we're doing right now. We're trying to understand all the people who are engineers in big company, how do they think about Segment versus marketers in small companies? And again, beyond personas.

And going back to measurement a little bit, how do you guys measure that more specifically?

Yes, well, of course, with segments, like, measurement is at the heart of all of what we do. We measure very aggressively. And so we track everything that's happening on the website, outside of the website, on, like, second-party data, like emails, how people interact with our emails, how people interact with us socially. I think something else that's that's happening right now, we're seeing more and more enrichments happening. When we have a signal from a customer or potential customer, we actually enrich that data with providers such as Clearbit, and we then get a better picture of the customer, like a 360-degree picture of the company they work for, the size of the company, and that helps us a lot personalize the experience. So I think enrichment is something that's really key to our success right now.

So maybe we can go into tactics a little bit. It'd be interesting to hear about how your organization is structured and who's accountable for CAC, one. And second, if you think about your marketing tech stack, what does that look like? And we can dig into that a little bit.

Sure, I'm accountable for CAC. I think these two guys might say the same. I've got a really great product marketer who's here, Vanitha, a really great, demand generation person, Alex, who's here. I've got a content manager, designer, and we actually run SDR out of India. So for the processes inbound leads, and that does report to marketing. In terms of technology stack, we use HubSpot. There's been the typical back and forth with the marketing org about Marketo, HubSpot. We're at HubSpot right now. Salesforce, standard acquisition methods, Google PPC. We use a few different lead providers. We've tried and had some success with early in the buying process providers like TechnologyAdvice. Later in the buying process, SoftwareAdvice we've used with a lot of success. Capterra PPC we've used with pretty good success. We're working with G2 Crowd. I think pretty highly of them. But the challenge with all of these is at the end of the day, the volume is at early in the buying process, and executing against that, the rest of the conversation brings down your cost of acquisition, and we're still a work in progress very much on that. But we know how to generate a lot of volume early in the buying process.

Yeah, I mean, I'll talk about OpenDNS, where I was there for 3 or so years, and at Cisco it's interesting, We are leveraging a lot of what we're talking about here at a very massive scale, but I think more relevant for this conversation is OpenDNS. We were a bit of point solution junkies in terms of marketing and sales point products, but they really drove a lot of decision making that we did. The core that we did, obviously we were a Marketo shop, we used Lattice engines, we used Gainsight, because I think again it's part of the the whole customer journey from brand introduction to renewal. And that's sort of when I think of the upsell motion, it's a big part of that. And those are the core tenets. But we were very much about trying out different tools to see how it did. And again, throwing spaghetti at the wall is okay, but you have to measure and then make sure it doesn't confuse your overall outlook. We also used BI tools. Really to look at again, not just marketing, but across revenue management, we called it. And that was a really successful way of doing it.

Yeah, so I'm definitely like the tool junkie. A fancier word would be like marketing technologist. It's the same thing. And so like we use a lot of tools. So we spoke about Clearbits. And again, that's all those tools like for B2B. Clearbits for like enrichment of customer profiles. We use MadKudu, which is a great vendor that we use for all of the scoring, scoring the lead quality. Just give you an idea of how performant that is, they're able to find 7— to predict 75% of our revenue within 10% of our leads. So that's pretty significant. We use Drift for the live chat, which has been like super successful for us in the past month because we use it in a pretty unique way. Customer.io for our emails. MonkeyLearn for the machine learning, a lot of Zapier here and there, and of course Segment.

And if you maybe to turn this around, when CAC is not going well, when you're getting not the results you want, how do you evaluate these tools? And when do you say, now you're out the door? And if you could share some stories, maybe one or two stories around that, that'd be—

Well, what you say is just don't sell to smaller customers, please. Just immediately cut out the small customer segment. CAC looks a lot better, right? You're only marketing to mid-market and large market. Go to mid-market. That's a quick solution to that problem. I'm not winning that argument, by the way, unfortunately. But look, we actually have— we sell to small businesses, mid-market, large market customers. So we're everywhere right now. And I'll just be honest. I mean, cost of acquisition to the small business segment has just gotten insanely high, almost impossible to get to a CAC of 1. And it's just increased competition. There's some implicit market things happening, obviously, with all the pay-per-click programs getting higher in price. And you can see that from Google data. The answer is invest more in content. But that's sort of longer term. And until we start seeing gains from that, it's a waiting game.

So if you push the content button a little bit, And you're the expert in content. How would you, what kind of advice would you give a company thinking about going to content as a first step?

Yeah, don't go half in, go all in. Go all in. And by all in, I don't mean just writing the content, right? All in is have all of your employees publishing on LinkedIn. Have a strategy around distribution for every single post you put out there, right? Have impactful posts that come from your CEO or whoever's appropriate in the organization. And do the gamey things, which is have your entire employee base like it on Medium, right? If you get to 50 Medium recommends pretty quickly, you have potential to get into top stories. And it can go very fast. There's gamey things you can do like that with LinkedIn as well and other gamey things you can do with SEO. But go all in. Make the full investment. Don't wait. Yeah, don't do it halfway.

Jeff, any horror stories where you were killed by CAC?

You know, I'd say every time we saw that coming, we actually changed strategy in a sense. So the first time at OpenDNS was when we were starting to go upmarket. We really embraced the free trial, right? Which was, you know, we were a SaaS security company, which is pretty hard to do for those personas. And we made a focus, a company-wide focus to build the free trial as part of the product. So the customer journey again began on the website and then went to, The call to action, every call to action, you can still see it when you go to the site, is for a free trial, and then we optimized that, and we brought in new technologies actually to measure that and to make sure the experience was meaningful for the free trial. Now, if you look at it, 75% of our ARR goes through, has a free trial. Some of them are fully deployed, so it actually serves a lot of things, but it allowed us to show the kind of value as we were making a transition, and CAC was going the wrong way. The second time this happened was as we were getting larger and we were, I guess, Series C round. That probably means different things to different people, but when we decided, listen, we need to focus, and that's when we started embracing some of the account-based marketing techniques where we would focus on the 1,000 companies that we wanted, which all of a sudden, that's like you're not going for the world, you're going specifics and getting innovative programs and content that spoke to the innovation of what we do and also how we brought them through. Those are two areas where we actually changed strategies Again, memory has an interesting way. There were some failures along the way in getting there, but that's how we ended up.

Going to the content marketing, or not the content marketing, the account-based marketing strategy, what kind of advice would you give startups who are thinking about going that direction?

Start small. It's kind of account-based marketing. It's focusing on your customers, and you can do it in a smaller segment, and you can do it with not a lot of money. I mean, you can have, you know, we did things like just sending a unique item and following it up. The other thing, and other people have written about this too, is like, it's really account-based marketing and sales, right? It's like, that's a really important tie-in, and it's a great way if you're having any sort of friction between marketing and sales, which I'm sure never happens, you know, it's a great way to bring them together, because it's very much aligned with sales goals. They have people, they have a funnel, and most of the time it's mid-funnel. And there's great examples of doing a very great account-based marketing campaign, which could be as simple as sending a picture of a— we did a whole visualization of companies' networks, and we would send it to them, and then that would start driving a new kind of conversation. And then the salesperson and the SDR write LinkedIn, and it can bring harmony.

Great.

And Guillaume, I know that segment is about experimentation. So how do you think about creating experiments When do you say this is successful or not? And how do you move on from that?

I think there's one thing that's interesting that I can talk about, which is like in B2B, the sales cycle is extremely long. We're talking of like anywhere between 3 months, 6 months, sometimes 9 months on like big accounts, right? And so like if you have a campaign or an experiment and you gotta wait 9 months to know if there's actually positive CAC, well, your speed or your pace of experimentation is extremely low. So that's a problem that we have solved recently by using one of the most recent API endpoints from Clearbit, which is like IP— reverse IP detection to find the domain behind an IP. And the coverage is pretty good. It's like around 30-40% of domains we're able to find. The good thing about that, it means that we can actually score those companies and understand is the traffic of high quality. And that's immediate. That takes us like 1 second to 1.5 seconds. Which means that instead of like going, waiting for 9 months, within like 1 second after the hit, we know if that's actually a potential lead. It leads to a lot of like important discoveries. The first one is after a week, you have enough traffic, you know if the campaign is gonna work or not. You can predict the result of the campaign. Second thing is that if you know it's a high-quality lead and it's hitting your website, should you do things differently? What can you do differently? One of the answers is of course to pop-up chat. The second answer is to change your website dynamically. In our case, for example, we're working on something to hide dynamically the pricing page and the sign-up button when there's a high-score visitor, because we don't want them to sign up. We want them to go for a demo, right? So it's the kind of things that we do.

That's great. How long did it take you guys to have this insight? You know, what was the evolution of this?

Um, I think— I mean, I get approached by vendors a lot for, like, new APIs and new endpoints, and I have, like, a whole funnel of, like, testing my experiments, Now we're building really on top of this whole like IP discovery and enrichment and scoring and adding a lot on top of it, and we're going pretty fast. I think the challenge we're finding right now is that we're seeing more and more signals that we can buy. People coming to websites, people attending a conference is a data feed we can buy also, and those tend to overlap. And so what should we email or what should we chat about with the people? And that's a huge problem, and I think the next frontier is going to be building a kind of an AI layer, which is going to say, inform us, who should we speak to within the company, with which channel, email, chat, website, about what topic? Because there's too much overlap. That's something we don't have yet, but we're really thinking about.

There's a lot of noise in the martech ecosystem, a lot of vendors promising, we're gonna help you lower CAC, we're We're going to do X, Y, and Z. What advice do you guys have for startups trying to reach you?

Good luck. No, I think it's hard. It's like the value proposition is so digestible, and you love it, and you want it to work. I mean, we've all been burned many times. But we also have great vendors that we use and love and trust. It's being able to show value and making sure implementation doesn't kill you. And those are two things, like really being able to show it in a short term, in a short period of time, and making sure that like the implementation doesn't require you to, you know, tap the CIO, the CFO, and everyone else to actually get it done because that's just, those are pretty high barriers.

Yeah, yeah, don't make my job more difficult. And there's just a lot of vendors out there. There are a lot of vendors trying the same approach. We're at a weird stage in demand gen where everybody has just converged on the same solutions to market to us. And that's a bad place to be. I think originality and authenticity are key. And yeah, people want to feel like there's a personal touch on everything these days. And I think a lot of startups have created a false sense of rigor around over-measurement of stuff and over-optimization, and it shows in their marketing pitches.

I'm very playful, so I like testing a lot of things.

I'm gonna hear from a lot of people right after this.

But there's two things. One, because I'm segmented, it's very easy for me to send my data. I mean, a click of a button, I just start sending my data. And so I never accept a pitch based on like fake data. I always ask companies to work out of my own data and show me actual results. And then I do Hunger Games, which means that I actually put startups that do the same thing once against the other, and I ask them to share the results so that I don't even need to discuss closing with the best company. They see it from themselves. I did that with the scoring technology. I actually had 3 vendors work out of the same database and write back the scoring to the same in the same warehouse, and it was super clear who was winning at the end.

Another quick thing that I thought of is like when you can connect it to the other parts of the go-to-market machine, you know, like that's what we're dealing with, right? And so, you know, and most of the data that you're providing and insights you're revealing connects there. The more you can build your product to help facilitate that for us, I think is a, Those are companies that will win, in my opinion.

The other thing about startups is people want to hear from the CEO. They don't want to hear from a BDR. They always want to hear from the CEO founder, right? And we can pretty quickly identify who's a startup and who's not. You may try to appear big, but we can vet that pretty quickly, right? There's Crunchbase, there's LinkedIn, there's whatever. And if you're a 10-person company or even a 20-person company and we don't hear from the CEO, I don't want to talk to you. Yeah.

So maybe the one last question, two questions. One is what is your biggest challenge with CAC? I know Sonny, you touched upon some of that. And second, what advice would you leave the audience around how do you improve their CAC? And so that they can also come back to you with thoughts or advice if they have some.

I gotta process my thoughts on this one. You seem to be on.

I'll give you counsel on things that no matter what size company you work at, especially if you're going into a new market or even your own, take some examples of your last 10 wins in the different markets and do a point-by-point analysis of what they did to get there. I'm shocked at how many companies don't do that. And you can learn and do it again with sales, with marketing, with customer service, however you do it, and map out what happened. What moved those customers to do it? Look for those trends. Look for where you were successful. What was the investment point? 'Cause again, the whole point of the metrics that go into CAC are where to invest more or where to stop investing. And it's amazing what you could do in finding out new areas of investment or where to stop investing. Places to pull out, but also what content worked. And sometimes talking to salespeople on what actually moved the customer to close, you can then take that and turn that content into a digital asset at the very beginning of the funnel and the beginning of the journey. So that's something that everyone can do. It doesn't require any money. It doesn't require a vendor. And the insights you'll gain from that I think are gonna be pretty tremendous.

How about an area where you would love to get some advice regarding CAC?

Yeah, I mean, I'm still looking for a lot of the predictive technology that can actually that can actually help me in a meaningful way. I mean, we've done elements and it works in the smaller, but especially at scale, like now we're talking about at Cisco. So, and also to be able to do it in multi-segments. We have, we sell to everyone and everyone is a customer. And so applying that, especially to the customer success journey, is something that we're always looking for.

I'd say get data, then get more data, and then start using that data to create a rich experience, a personalized experience. You'd be shocked to know, like, by how much a rich personalized experience has improved all conversion rates, which of course impacts CAC in very much the same way. Yeah, I mean, that's, that's my main advice. Go into personalization heavily. The other thing is go and be first into a new channel. The Cost of acquisition is so low when you're first to a strategy, first to market, it's shocking. So for example, right now Quora is opening like their ads platform. Go there.

How about area you're still working on?

Yeah, I mean, I'm still working on applying personalization into web on first page load. It's pretty easy for us to apply personalization on things that are not synchronous, like email, like live chat, phone calls, because we have the data. Using that on the very first page of an anonymous visitor requires processing that data at scale and at speed in, like, less than 0.3 seconds. That's technically very challenging.

And Sunil?

Yeah, driving CAC down, slash, my major takeaway would be find large swaths of people. Don't, you know, don't go after, you know, super target all the time. An example of that would be— your buyers all talk to each other. I'm sure you know that. But when we were building Scripted, two-sided marketplace, writers, businesses, and we needed to build up the writer side of the marketplace initially, we found that all of— like a ton of writers hung out on Meetup. And we found a loophole in Meetup, which was if you join a Meetup group and you send an email to the Meetup group, it actually goes out as if it's a personal email to the entire group. So you do that and you can find thousands and tens of thousands of writers very fast if you hire BDRs through oDesk or whatever. But large swaths of people. Your buyers all talk to each other. They hang out with each other. They influence each other's decision-making power. By getting to them at once, you can drive down CAC pretty rapidly.

Well, great. Thank you so much, guys. Please join me in thanking the panelists here. I hope you guys learned something interesting, and, you know, we're all still working on lowering our CAC.

Yeah, thanks.

Thank you.

All right.