Paid Ads For Marketplaces Workshop

So welcome to our Everything Marketplaces workshop, which is going to be on paid ads. So we've had lots of founders request a workshop on how to use paid ads and marketing when starting a marketplace. So we're going to be covering that in detail today. So Ricardo Manes will be leading the workshop. Most of you probably know him from his deep dive posts and breakdowns on paid ads that he's shared in the community. But if not, he has some really awesome experience working with marketplaces on paid ad campaigns. So as a quick note on the format, our workshops are once again 20 to 25 minute presentations that will be followed by a group Q&A. So if you have any questions throughout the workshop, you can post them in the Zoom chat. Just be sure to turn on your camera and you can use the Zoom raise hand functionality so that way we can call on you when we get to the Q&A after the presentation. So that said, you know, big thanks for joining us here today to lead the workshop, Ricardo, and I'll let you start with the brief intro and just take it from here.

So hello to everyone. Um, who I am, just a brief presentation. So I'm like Ricardo Montes, um, I'm founder of like DealShake, a marketing agency specializing on marketplace for helping them to scale with pay ads. My background, I also founded my own marketplace here in London, UK, a marketplace of like independent retail and also scaling using like pay ads. A little bit more details about today on the workshop, a couple of things we're going through is How to combine SEO with Google Ads, because one of the most asked questions is how to transition also from SEO to paid ads. And also a little bit of deep down on like Google Ads, some example, something also in terms of like campaign structure analysis, where to start thinking about also allocation of budget. Google Shopping Ads is any of you also is going to use it. And then a little bit details of like Meta, how a funnel, full funnel structure of Meta campaign look like. A little bit more example of the top marketplace. Also here they're using like a Meta Ads and some recommendation. So hopefully you can run out of like this workshop with some insight if you're running ads or at least planning to running ads in the future. So the first thing is a lot of marketplace founders ask me, what is like the right marketing strategy? Or like, which channel should I start with my pay ads? Should I start with Google? Should I start with Meta? Should I start on like TikTok? Which is like the best? So first thing first, there is no like best for like every single marketplace. It's very dependent on like your marketplace and also your audience. For example, are you focusing on like acquiring demand, supply, where those people hang out mostly? It's like, as I said, Google, they're looking for like the supply from the demand side. It's like Meta, where those people stay and also like metrics, for example, Lifetime value and AOV, and also take rates. And also, what is your objectives for like pay ads? Is for example like having extra cash flow and having profitability? Is more like grabbing market share here? So it can be very depends on case to case, but also there are similarities. And then other way to test, there are like 2 approaches. One is like sequential testing and one is parallel testing when you want to test different channels. So here depends on like sequential testing, you testing one channel at a time. And other question sometimes is like, how long should I test a channel until I know it's working or it's not working? It's very hard to say. Depends on the budget, depends also like transaction, the amount of conversion you get there. But as I say, normally can be a little less, like, uh, 3 to like 5 months sometime. But it very depends on your budget and also your audience. Parallel testing, you have a little bit more budget here, so you can test in parallel multiple channels. For example, like Google, Facebook, TikTok at the same time. The advantage parallel channel is quicker to getting data and learning which channel could work, but also is much more expensive rather than sequential that you can have a small budget and testing channel by channel. In terms of like SEO, this one is like one of the biggest questions from founders is how to move from like SEO to Google Ads. So if you have like SEO working for your marketplace, potentially Google Google Ads can be like a very good way, good channel to also like moving and increase your acquisition. And why? Because as I say, you already have a lot of data in terms of like query pages, what is also like driving some conversion. Of course, majority of your SEO can be also like in terms of like of the funnel, or maybe you have split your service by category, by city, but then focusing on specific also keyword searching inside like Search Console that can give you like a gap in terms of like which keywords also you can use on your Google Ads to pushing even further your SEO expansion, or also understanding what is the overlap between the two. So they can help each other out. And a couple of things is like how to combine them. As I say, it very depends on your data, your SEO data. The main thing is like you want to understand also which is the synergy you can have for them. For example, if you have specific keywords that are like driving conversion and you're ranking, starting to rank them organically, having a specific like campaigns, specific ads pushing them further can be a way. So using basically Google Ads to boost SEO. Other thing also can be maybe if you are like at the very start, I saw few marketplace using Google Ads also to test which keywords can work well and then going back and creating the SEO. This one is the strategy. So from like Google Ads to SEO, the main thing about this strategy is you can use, as I say, budgets on Google Ads to testing the ground on specific keywords and understanding also before you starting your programmatic SEO maybe which keywords that are the most interesting or they have like the best, the better performance so that you can construct also part of your programmatic SEO based on those data. So having like Google Ads leading your SEO. The main things about this strategy, as I say, you can save a little bit of time on gathering data to build on top of your SEO. But then it's also, as I say, you need time and budget as well to like having enough clicks, enough impressions on your account or all the different keywords you want to gathering data to build your SEO strategy on top. And then some deep down on Google Ads. So here we have a couple of examples of 2 very famous marketplaces, Outdoorsy and RVshare. And let's have a look at the Google Ads. So here on Outdoorsy, I classify those in like 2 categories, low intent and high intent. What I mean here, for example, low intent are more like a generic category. For example, we can see here for Outdoorsy, there is like they promote the overall marketplace. They say Outdoorsy official site, number one trusted site for RV rental, Outdoorsy RV rental. And then we have the high intent, is very specific. So in this case, it's like bottom of the funnel when it's more like keywords, the leading, uh, conversion. For example, in this case we have Detroit campers and RV rent outdoorsy. Also here I found some like dynamic insertion. So one thing you can do is also in your Google Ads having dynamic insertion on keywords, also location, so that based on like the location of your user, they're going to see the name of the location in the ad copy. And here the main difference is this, with a low intent is more like try to having like on top of the funnel and then what is important. And then on high intent is like pushing on conversion. Other example, quite similar, low intent, high intent. Here for RV Share, we can see the low intent is more generic, like RV rental near you. And also, I guess this one is also like a very popular SEO keyword for them. And the high intent is specific on the city. Couple of other things. So here we have How to start? You need to thinking about like what is like your high intent, a low intent. In this case, for example, from your demand side, but also those company running ads also for supply side. Other thing that is quite interesting is like the type of like objectives they have for low intent. For example, objectives for low intent is to bring awareness. To letting people know what is the most important things on the platform and the platform, the value of the platform. The high intent is helping normally, as I say, supply or demand side to doing a specific action. As I say, in this case, like, can be a conversion, can be a registration, and so on. And here's some other example about budget allocation. As I said, those are just like generic examples. So don't take those percentages that can work— those percentages that can't work for every single marketplace. But just to give you like an idea, normally when you have like early stage budget allocation, you try to focus your budget— if you can see on the bottom of the screen, on the bottom of the funnel, like 70-80% of the funnel Why? Because you want to capture, as I say, as many transactions as possible, as many conversions as possible. And then also you have like small budgets for like middle of the funnel and top of the funnel just to getting enough people also to the bottom of the funnel. So to starting having people also experiencing and understanding about your service before to convert, because not everyone as I said, on the first touchpoint is going to convert. And difference with more mature marketplace, the budget allocation is shifted. So in this case, the majority of budget is on the top of the funnel. And the reason why is because they try, as I say, to expand the market, the market size, as I say, making them the marketplace more mainstream. So they're investing even more on the awareness stage rather than they have also like middle of the funnel and also bottom of the funnel. They're working together. And this one can be also the case, for example, like Airbnb. They are like very, very focused on the top of the funnel, try to bring more people inside the platform, but also having bottom of the funnel Like re-engagement campaigns or very like specific high-intent keywords. They're very transactional. And this one is like the main difference, just to sum it up. So early stage marketplace maximize the bottom of the funnel. So they, all the keywords, like long-tail keywords, they're like very specific for purchase intent. And then having also some budget called top and middle, to having a pipeline of users that are starting, as I say, understanding your marketplace and then moving along to your bottom of the funnel. Scale maturity is more like, as I say, top and middle of the funnel, but also you want to maintain some allocation on your bottom of the funnel. Here are a couple of more examples, also like StockX, One example here on low intent over high intent. On the high intent, we have circle here in blue, like the shopping campaigns. So in this case, they have a lot of like shopping campaigns, like more than 90% of the ads are like shopping campaigns. And here I just want to give you a couple of like insights on shopping campaigns. So this one was some data. For one marketplace I was working with. And one thing we recognize is also having— using a performance bucketing strategy, understanding also classify all the different products based on different type of ROAS. So they're going, for example, having the index is like average ROAS, like lower than average, higher than average, and then understanding How much is also like the cost? For example, here we saw that the budget, 97% of the revenue come from like high performance product, but only 54% of the ad spend goes to the went to underperformance. So in this case, with like performance bucketing, we saw that a lot of like budget that we're spending with products that have like Very very low ROAS, like zero or one percent. So they were like losing a lot of money. This one is an example. Under index means it's like low ROAS, for example zero one ROAS, and they were losing more than like hundred grand in profit. So this one is not like revenue GMV. This one is like profit. Accounting for like the take rate, and also in the last 90 days there was like 24%. So this one is just one strategy when you're also starting to be more advanced, and if you're running shopping campaign, to fragment all your products based on like performance bucketing and understanding how those products performing and where the product is, the profit is coming from. And also here, just to let you know, a little bit more advanced on this, also classification of product based on like number of conversion and also performance. So for example, products that normally they don't have enough impressions, we call zombies. So because we don't have enough data to understand in which step in terms of like ROAS, they're going. And then we have profit products, they are like flux. Maybe they have like 1, 2 conversions, but they have very high performance. But we can't say if that is the case in terms of understanding, is that going to be driven a lot of like purchase in the future? And then we have products that are costly and profitable. So costly, they have basically, as I say, many conversions but cost too much. Profitable, they are like the best products. And this one on the left-hand side is like a type of structure of different campaigns where the heroes basically are like the top-performing ROAS and index ROAS. And then we have some branding and catch-all. This one is basically having low percentage of budget and having across all the different catalog so that we can understand if there is any other products that are performing, new products. And then we have villains that basically like the Flux and the Mesh, the one like, and the Costi, the one that basically are not profitable and they are not driving lots of like profit for the company. And then we have the zone-based. So basically here we have a catalog of products that we try to get more data to classify them in the different profitability buckets. And here also other questions sometimes is regarding like Performance Max. Does Performance Max is like the best type of campaign to start or not? Normally it's not because Performance Max is working very well when you have a lot of data in your account. For example, when you have implemented server conversion and you have a lot of number of conversions in your account. I put some pros and cons. The pros is that you can test different networks, for example, search, display, video. Cons also here is that you need a lot of those data, for example, between 50 to 100 conversions per month. Google recommends 30 conversions in 30 days, but normally it's not that enough. So normally this one, Performance Max, is more recommended if you have a lot of data in your account. Another thing that is recommended is using a script. So this one is an example of like a campaign I was analyzing for one marketplace where Performance Max was spending in 3 different networks: video, display, and search, but the conversion came only on search. So Google doesn't tell you. It's a little bit tricky how we say here in London which network is bringing your conversion for the Performance Max. But as I say, using a script, you can analyze all those data. So a little bit more advanced. So if you want also using Performance Max, you have enough data in your account. I suggest using scripts for analyzing all this data that Google doesn't give you directly. And then a couple of common pitfalls and how to fix when you're starting with your ads. The most common thing I see when I do audit of marketplace account and the Google Ads is problems regarding like clicks and budgets. For example, a lot of, a lot of you, if you're running like Google Ads, can see your campaign is limited by budgets and Google say, hey, increase your budgets. But the main thing here, apart of like increasing your budget, is understanding how many clicks are you getting out of your budget. Because in the majority of the case, like also excluding keywords that are very costly, or also work out a maximum CPC for your keywords, can be a good way to understand if you want to be profitable with your ads, specific keywords that you are— as I say, they're not going to work for you and make your ads profitable. And then other thing is limited by impressions. So another thing is like the quality of your account, and basically Google decides how many impressions your account is going to receive. And that means if you have low impressions, a problem is that not many people see your ads. So a way to work it out here normally is like increasing the quality of your ads. So the Quality Score, relevancy, having very specific category or ad groups, they're also pointing maybe into the, your category page, so you can have a higher relevancy also on the ads and try to get more impressions. Other thing is, um, few keywords spend all my budget. This is something that I see I saw quite often. So accounts, they may be having like small budgets and the majority of like spend is around maybe 20, 30 keywords. They're spending all the budget and they're getting like really few clicks per week. So my recommendation, if you try to make your ads profitable, as I say, not bidding on those at the early stage of your account. As I say, another thing, you can have like a target, as I say, a maximum click. So you can put like how much you want to spend per click, but in the majority of the case, those keywords are not going to work for you early on in your account. Other thing is like optimization of your budget and also the keywords. Most important thing here is understanding and doing a very good keyword research. For example, here you can see Google in Keyword Planner giving you like the top page bid, the lower, the top and the lower bids for a specific keyword. So it's very important to understand also what is realistic. Because sometimes you have like keywords in your account, they're like very expensive, the range of bidding. And normally for your daily budget, you can get, as I say, 1 to 2 clicks. So those are keywords that are not working for you. So try to stay away, especially if you try— if you have a small budget, and also if you're looking to have your Other thing here also in terms of like optimization, sometimes problem can be related to with a small budget, try to target a very wide set of country or also states if you're targeting US. So a good approach is also try to be data-driven. So looking in your account and understanding like which city which states they're working best for you in terms of volume of conversion, but also cost per conversion. And similar to this is also adjusting your bid strategy and also your ad schedule. For example, you can check your ads and see if there are specific times in the day or specific days in the week you have a cheaper cost per conversion or big volume. So that you can start to exclude also time and days that are not working for you. Other thing about bid strategy, when you're starting an account fresh, my recommendation is just having— try to have as many clicks as possible. Normally you go with, like, as I say, maximize click, max CPC. And then when you're starting to have enough conversion data, you're starting moving along with more like a smart bidding. So don't start if you have a fresh account just with like smart bidding, like target conversion value or target ROAS. It's just because the account doesn't have enough data to help you, giving you those conversions. And here also other things to notice is how to improve You add copy, very, very important. Normally when I do audit, I see that a lot of like copy are very similar. So thinking about also what are like the benefits from your user, if you're targeting, for example, like the demand side, you're running ads for the demand side. And the best approach, a secret approach, is like speaking with your user, understanding exactly what are the benefits for them. And using those exact words in your ads and also in your landing page. That is like the secret sauce to have very good click-through rates and converting ads. As I said, don't try to reinvent the wheel. Speak with the user, getting the data, understanding exactly how and also which keywords they use And then you have those in your ads and landing page. Other thing is also like do testing. Everything is based on data, so doing a lot of like A/B tests can be very important to understanding like geography, understanding also like what is working best for you. So that can be especially in the initial phase to put in aside percentage of your budget, normally like 10 or 30%, for like running specific tests and trying to get all this information of like new geography or like new category and so on. Um, in terms of like meta ads, this one is how a full funnel campaign structure can look like. So we have campaign for example, from the bottom of the funnel, just on the bottom. Those are like the main type of campaign. I like conversion here, especially if you have, as I say, early stage marketplace you want to tackle as conversion for early on. So you don't want to have like click. The only exception here we have like traffic campaign. On the top of the funnel, it's just because that can bring more, as I say, cheap traffic to the website. But then you need to assess if the traffic is relevant or not. How to assess? Little trick understanding is how long a user spends on the website. If they're spending, for example, your average time session on the website, that can be good traffic. And then we have campaign for like middle of the funnel, interest-based, Also, we have a middle of the funnel Advantage+. Advantage+ is a campaign that Meta is optimizing like some of the components for you, but also you need a little bit more data sometime in the account to make those work. And other thing is Meta Ads catalog and sets. So inside your catalog And the catalog is basically all your products, similar to the Google Shopping, just inside Meta. One thing you can also create is you can create a subgroup of your catalog that's called sets. And my recommendation is also testing what are, for example, like the top-rated product, because by running a lot of ads, I see the majority of like users sometimes buy products they already have a lot of reviews. Or as I say, if they have a specific, specific seasonality, or if they are like specific category, they are more interested as well. So a good, a good rule here is having like a different set of your catalog and that you can run for seasonality, you can run by category So that you have a more granular level of data on the performance of those specific catalogs rather than running all the catalog and spending money on all your products at the same time. Here, an example of stock X creatives. For example, in this case, only 11% of the Meta ads are video. However, we can see a lot of big trends of marketplaces using more and more video. And also they, in terms of like the message, they try to underline the benefit of like verified sneakers and verified sellers. And also they're using a lot of UCG. Other example, we have TamTak. The creatives, I say all of the creatives here, they were like videos. So for them it's like performing quite well. And then the main thing is also having different campaign for different stage of the funnel. And also they try to using UCG as well to showing the product. Fresh Air, also here we have 89% of Meta ads, they are like video. So video, they're working quite well, especially in terms of like click-through rate, because that is the first goal. Try to get like the click, try to get people landing to your website, and then try to convert those. And also they have inside the message very specific, very specific copy. For example, subscription-free booking software. And also they try to highlight all the main features that they have and benefits. Other thing about testing, this one is another question that I receive quite frequently, how we can test ads on Meta. There is a lot of different ways to do it. This one can be a good way to start so that you have, for example, 2 different campaigns So first campaign is like a testing box. So basically where you test different creative, different message, different audience. And then when you find something that is work, you create a second campaign where you have only the creatives that is working pretty well and you try to scale. So in terms of like conversion, of course, also having like a campaign for like remarketing. Is very important. So just reminding people about, about your company, and also especially because they already engage with your company. It can be, as I say, a website visit, can be also starting like part of like the, the checkout journey but didn't finish. And also here an example on, on the right inside. Be very organized on your naming your ads when you're like testing. For example, here we were testing like different creatives and we have the same message and we were only changing one variable at a time. Don't change too many things at the same time, otherwise you are not sure what is going to work. It's like the image, it's like the audience, it's going to be the message and so on. And also here, a little bit of strategy to increase conversion. One thing is also thinking about how you can shorten your marketplace. So this one is like a TEAL marketplace for teachers and students to learn guitar. One thing also they were like testing is like to try to understand if closing the marketplace or leaving the marketplace open can convert better. So what does it mean? Closed marketplace, as I say, you matching specifically, you can't let the user browsing your supply sites. Open marketplace, you have all the filters, they can browse all the catalog and so on. So that can be also a way for thinking about how you shorten the user journey. And normally having like a match can be a good way, but it very depends on like different marketplace. Other thing is also using like AI, more and more marketplace using like AI to matching the supply and the demand. Why? For shortening the user journey. And this one I just took from like a Growth Mentor that is like a marketplace of, for like tech mentors. And here basically rather than searching all the catalog, you just adding a couple of reply to a couple of question and the AI giving you automatically 3 to 5 match. So that can be another way also to experiment. And lastly is like GetYourGuide, time scarcity. I think they're doing pretty well. And basically it's a platform that you can book experience. And once you're adding them to the cart, they're also starting like a timer. For example, you have this reserved for 20 minutes. And based on this, they can also try to adding scarcity So that people try, as I say, to go back and also you receive nudge on emails as well. So that can be also something that you put on your ads, for example, like specific limited quantity available and so on. And those are like some good example I found. And now we have the group Q&A.

Awesome. Super helpful. So I appreciate walking us through and you packed in lots of insights there. So I know it's going to be a lot for us to kind of digest here. I actually had a question in the chat and it was, do you have like a recommendation for the minimum budget and maybe duration for running ads and initially kind of building up one or both sides for a B2B marketplace?

Yeah, that's a good point. So main thing here is understanding the channel you want to start. For example, if you're thinking about starting for like Google Ads, My recommendation is doing some like basic keyword research and try to understand what is like average price of high-intent keywords. Because sometimes if you have like average price very high, like $10, $20 per click, you can do your maths with your average conversion rate and then understanding what is a minimum budget to have at least a bunch of conversion.

Yeah. Is there like a rough kind of like amount for that? Maybe also too, like with the duration or?

As I say, it's very hard because some marketplace, maybe they compete in some location. As I say, the click-through rate, the cost per keywords are very high. So as I say, my advice is doing some maths. But as I say, if you want like something can be having at least maybe like 30, 40 conversions per month. So maybe like something around like $5,000 could be a good starting point, $3,000 to $5,000. But that can change based on like the price for keywords.

Yeah, no, it's gonna be helpful. I know it depends, right?

Yes.

Cool. Dmitry, you wanna jump on?

Yeah, I have a question for you about how do you do kind of the cost-benefit analysis where you are using marketing to Attract a user to your site. So you were suggesting to compare kind of the cost of that, let's say, click multiplied by your conversion rate and kind of the value of that conversion. But I'm worried that that misses the fact that if you, let's say, you bring an additional buyer to your site, that also makes it more attractive to additional sellers. And how do you incorporate not just kind of the immediate, let's say, LTV increase, but also that attraction sellers, which makes the site more valuable to buyers and kind of how do you incorporate all that into your cost benefit analysis when deciding how much to bid, what your budgets are and so on?

No, I think great question. And the main thing is also like if you have some details about the LTV of both sides can be a good start. And thinking about also, for example, what is the ratio, for example, between supply side to demand side for having a supply. So if you have those data, can be great to thinking about and say, okay, maybe like an acquisition of the user, you generate a specific LTV for the platform. And so you can go back and calculating what can be like your maximum cost per acquisition. In terms of like practicality, of course, like the marketplace, more like people coming on supply-demand side and more valuable is like the market. And because it's generating more value for both sides, that is very hard to like, as I say, track. One thing is like metric, for example, how many, as I said, if you are like demand side, how many supply side do you check before an average conversion? So that can be something that you're starting to analyzing and collect so that you can starting to have like, as I say, some, some data and thinking about, okay, every single like supplier bring, I can increase also like transaction by X. So you can think about maybe we have 10%, 20% more room in the LTV. So it's something that you're starting to reflect on to expand your CAC.

That's a great question. Hey, Ricardo, I actually had a more, you know, kind of basic question, another one for you, but it's a common question from founders when first getting started kind of testing out ads. Would you say that this is something that's typically kind of like founder-led? And then, you know, how do you think about like when the right time to potentially hire for this is?

It's a good point. I think it's very dependent also on cases. For example, like marketplace, they maybe have like raised more capitals and they're looking for having specific goals, normally like 18 months. Normally they have less room for the founders to learn the rope and they have also the budget to hire someone early on. If it's not the case, I think, yeah, starting for the founder starting the ads can be great because also you're starting to have some basic learning and some basic knowledge that can help you also understanding if you're hiring someone, how to collaborate, what is the priority for them, and also choosing someone that can be good or not.

Cool. Well, I think we'll wrap it up on this. I know we covered quite a bit and packed it into the presentation. So I appreciate walking us through. And as I mentioned, kind of before we started recording, you know, it was a surprise we actually haven't done a workshop on paid ads yet. So this will be a good kind of a deep dive for us. And, you know, we can always expand on it in the future and also continue the conversation in the community. So we appreciate taking the time to join us here, Ricardo. And thanks everyone for the, for the questions.

Thank you.