In Reviews We Trust

Ep. 10: How to mitigate the cost of click fraud across all channels of paid acquisition: Neil Andrew

July 20, 2022 Callum Mckeefery Season 1 Episode 10
In Reviews We Trust
Ep. 10: How to mitigate the cost of click fraud across all channels of paid acquisition: Neil Andrew
Show Notes Transcript

In this episode, host Callum McKeefery - CEO of REVIEWS.io is joined by Neil Andrew, founder and CEO of Lunio (formerly PPC Protect).  They discuss the changes in demands of VC's for capital fundraising, true cost of click fraud across all forms & channels of paid acquisition, how to mitigate these costs, the risks of using audience expansion tools, how best to manage refund requests and tips on the most effective way to manage your paid campaigns.

PLEASE NOTE: At the time of recording, Neil's company was called, and was referred to during the interview as PPC Protect. This company has been rebranded to Lunio since then.

Episode links:

Neil Andrew:

LinkedIn

Callum McKeefery:
LinkedIn
Twitter

Show notes:
Lunio (formerly PPC Protect)
Confused.com
Boohoo
Triple Whale
Shopify
Nike
Europcar

REVIEWS.io is a leading online review platform. It's used by over 8,200+ fast-growing brands. To find out more please go to:

www.reviews.io
Find us on LinkedIn

Callum: So today on 'In Reviews We Trust Podcast' we have a very good friend of mine, Neil, Andrew, Neil's been, a long-time veteran of the industry. He started, PPC protect several years ago and it's grown to become one of the number one PPC fraud detection services, so you don't have competitors and unsavoury characters clicking your PPC ads. Thank you, Neil, for coming on. I know we do talk on LinkedIn and on Twitter.

It's just great to have you actually finally on the show. 

Neil: Thanks very much for having me good to be here. 

Callum: So can you give the listeners, I know I just gave them a little bit of an overview, but I imagine your description of the business is gonna be much better than mine. Can you give 'em a quick overview of what PPC protect does?

Neil: Sure. So we're, a SaaS platform, so software as a service, working invalid traffic detection. So essentially we work with your paid acquisition channels. Be that Google Facebook, Microsoft, Twitter LinkedIn, Pinterest, TikTok, [00:01:00] anything you can imagine. We work with those channels to make sure the traffic that they are sending you is legitimate users with the real purchase or real conversion intent, and we automatically prevent things like you mentioned their competitors clicking on your ads, bots and scrapers, ad fraud and malicious web activity. What we find is depending on the channel, anything from 7% to about 20% of what you're paying for is invalid activity. It's not the type for the activity you want.

It's not real users, it's automated tools and so on. So we prevent that moving forward, and we help marketers make their ad budgets go further. We make the data they're working with cleaner. We make sure the analytics and BI tools they're working with are populated with, you know, clean, genuine data from real intent users and just make sure you can make the right decisions as well as obviously the cost savings associated with not having to pay for that kind of activity.

Callum: Yeah. So how, how does billing work with PPC protect? I, I do know this, but I want you to tell us because your description would be way better. 

Neil: Yeah, so we, we have a couple of, a couple of [00:02:00] different setups. Depends on, you know, client needs and the platform you're working on, but at a basic level, we take a small percentage of the amount that you've spent.

4% somewhere around there. And then we use that to essentially act as a protection fee. That's what we will bill you. And then, you know, you get an average 10 X ROI is what we target some clients of 15 to 20 X ROI. And then we also make sure that you know, if there isn't an ROI there, you're not making any return on that investment.

We actually don't charge you. So, so it was in our incentive to make sure that yeah, you are actually getting a benefit out of the platform. 

Callum: Yeah, I like, I like that billing method where you, you know, you could've, sort of backing up. What you're saying. So if, if it's not working for you, don't pay this. 

Neil: Yeah, exactly.

Yeah. It's one of our core values is honesty and integrity, which I know a lot of businesses say, but we actually genuinely try and live by it. Yeah. We're not in the business of charging someone for something they don't need. 

Callum: No, I love that. That's exactly like we are at reviews. If it's not working for you, you're not tied in, please, you know, feel free [00:03:00] to go and try, you know, spend your budget elsewhere.

You know, we're all about helping. You know, eCommerce brands become more profitable and if you know our solution's not right. And it's like, what you are saying if your solution's not right for them to shift. Yeah. And that's awesome. That's awesome. So, PPC Protect have you raised, have you raised capital.

Neil: We have. Yeah, so we did a pre-seed round in, late 2019. We did a seed round in late 2020. It was about $3 million around there that we raised and then can't say too much, but we actually just... we've just closed out a series A, so that's quite a, quite a significant raise, but I can't say much more than that at this point.

Unfortunately, I don't think I shout that. So yeah. Keep your eyes open over the next few weeks, I guess. And see...

Callum: that's really cool. Did it take long to do the race? 

Neil: Oh, okay. You have no idea. Yeah, I think we we're, speaking to actually. One of, one of the investors we talked to that we ended up working with and they said [00:04:00] "probably couldn't have picked the worst time in the last 10 years, to try of tried and raised as a tech company."

 So right. Yeah, every day the sort of public multiples of stocks are going down and down and down and yeah, a lot.

Callum: It's just getting worse. It's just getting worse. 

Neil: It, it is, it. It is. I mean, we, we're looking at you, we've raised a significant amount. We've got a fantastic valuation on the business.

We've got investors that we're very culturally aligned with, which is something that's perhaps not important to a lot of, you know, a lot of companies like there, but for us, it's super, super important above all else. The offer we ended up taking wasn't the highest valuation on the business. It wasn't the largest amount of money, but it was the best fit.

So overall I think we, we did well. Yeah, we have rapid, you know, over three X, year on year growth, which is like best in class in, you know, VC backed SaaS startup. So we're doing good on that side, but I think, yeah, we're in a fortunate position where we went out to raise capital at the point where we were profitable, which is quite unusual, I'd say in the space.

So yeah, we, we kind of had the pick of choices if we were burning [00:05:00] high, it would've been yeah. A lot more difficult. 

Callum: Yeah. Yeah. I, you know, I, I watch you guys and you kinda run in a similar, you, even though you've raised you, you sort of run still as kind of that bootstrap mentality where you are, you guys, you know, the team still hustling and they're still, 

Neil: Yeah, we always, we always run lean, you know, we have a big team.

I think for our, for our sort of size, there's 32 of us now when we raised our seed round was eight. So yeah, we've grown quite rapidly there. Yeah. But yeah, we, we've always been focused on this capital efficiency. Like my personal mindset is that. This VC game of, you know, how much money can we throw and see what sticks and burn millions a month.

I think it's nonsense. It's unsustainable, you know, stock charts can't go up into the right forever. Like something has to change as we're kind of seeing at the moment. So we always focus. Yeah, definitely. Tell me about it in my portfolios, in the gutter. But anyway, we, we always focus on getting that, you know, revenue to the point where it's like revenue is roughly equal or greater than the amount we've raised in total. And then we go out to raise again, rather than saying we raise a series a and [00:06:00] six months later, we raise a series B like that's not sustainable. 

Callum: Yeah, it, I, I think, I think what's, what are we in? We're in may 20, 22 and I think the tide is turning. I think, you know, it it's quite clear that there is a complete shift in mindset. I mean, I speak to VC's, probably on a, probably a couple of times a week, different VCs different calls.

And you're definitely seeing that shift in, Mindset looking way more focused on profit, which is that that's great for you guys, cuz you guys are doing that, but we're definitely seeing a, a shift and it, it's good. I think it, I think it's good. I think it'll it'll stop. I think they got used to the Uber model of kind of financing growth, you know, throwing money, throwing money at things which, which, didn't work well, not they didn't work, but they, they subsidized the growth.

They, you know, every Uber drive you took, you know, in the last couple of years has been subsidized. It [00:07:00] costs them to take for you to take the trip, but it's got them footprint. But I think even looking, I was listening to the call from the Uber CEO. That's all stopping. Mm-hmm they're they're gonna, you know, they're gonna work on profitability now.

And I think a lot of. Tech companies are gonna have to do that, else they're gonna get killed. 

Neil: I agree. Yeah, there there's. I mean, there is a lot of money out there, still a lot of dry powder as the VCs say to deploy, but what we've seen at least is those VCs are doubling down with their existing portfolio companies who need money to finance this growth, and can't raise externally because everyone is investing in their own portfolios now. So it's kind of a catch 22, a little bit there, but if we looked you two years ago and we went out and said, yeah, we're profitable and you we're raising. Our seed was actually quite hard to raise because VC's looked at us and went, what do you mean you're profitable? You're mad. You're a tech company. Like it's seed round. Yeah. That's, that's a negative. We don't want you to be profitable. Whereas this time around, it's been very much, you know, oh, thank God. Somebody out there who is profitable, like we're interested. [00:08:00] So yeah. Definitely shift in mindset.

Callum: Yeah. There's definitely that that shift that's happened, over the last weeks, weeks even I've seen it, you know, weeks conversations have just turned. So. Going back to the, the PPC protect solution. . Obviously for me, I've been hit by fraud. One of our competitors clicking our ads and that, that cost us probably about 20,000 pound over a three year period we didn't know it was happening. And then we found out about it, then we investigated and we did all the logs and everything else. We found all these clicks that they were, they were clicking on and, and it was painful, painful to see that you've lost all that money. So that was on Google. And you do also, do I know you said Microsoft there, but Facebook obviously is a big, big PPC paid media platform. How do you work with those guys? 

Neil: Yeah. So we work on a very similar sort of basis. The [00:09:00] actual sort of prevention methodology at the technical level is, is a bit different on Facebook than it is to Google and Microsoft. But functionally the same, like snippet, it goes on the website, just code snippet and tag manager or any tag manager you prefer.

And then we just analyze the traffic coming from Facebook. And when we determine that those users are invalid, we add them to exclusion polls. They exclude them against your campaigns so that you can't show any campaigns to those specific users. They can't interact with them. The difference that we see is what you say about like competitor activity, which I remember you posting about on LinkedIn.

Actually, that is, I was somewhat surprised, but also not shocked in the slightest, but we find that competitor activity is a small subset on Google. It's not the core of what we do, but when it does hit, it is the impact is significant, but we don't really see it on the social platforms because the targeting is very different.

You might get remarketed if you've been in the site, but it's, it's still very different

Callum: Harder to hit, definitely.

Neil: Definitely. Yeah. And what we find on the social platforms is Facebook. Especially a lot of [00:10:00] the invalid activity we try actually comes from fake accounts, like bot accounts, and obviously bot accounts, try to look human in some way.

And one of the ways, one of many is they try and interact with sponsor posts, advertisements at the same frequency and type that a genuine user would to try and trick, you know, Facebook's detection, algorithms, and Facebook don't do a bad job at removing fake users and fake bots. Yeah. There's about 15 billion accounts a year that they remove.

And it's twice the world's population that they remove every year in fake accounts. Wow. But they're just so prevalent that we still find out 22, 23% of traffic from Facebook is coming from these invalid sources. It's just a battle. They can't win. It's nonstop. So different to on the Google side. Yeah.

Not so competitive focused. But an interesting challenge nonetheless.

Callum: What about Twitter? Do you do Twitter? 

Neil: We do, we do Twitter. Twitter is, you know, from a marketer's point of view, putting my marketing hat on, pretty rubbish as an ad channel, I would say it's, it's one of those that you're not gonna get any direct attributable revenue from it.

You could argue brand uplift [00:11:00] and recall, and that sort of stuff. Yeah. But difficult to model. But yeah, we, we find Twitter has a, I mean, as everyone knows, if they follow the whole Elon Musk story. Twitter has a serious bot problem. Yes. Very serious bot problem. 

Callum: I, I think he can solve it. I think it can be solved quite... I don't know it, all of this bot problem for me seems to be some, you know, just to get verified customers on there. Why are we not, why are we not doing something like a credit card check or a a, yeah. I mean, some kind of check to just make sure they don't, they can be anonymous when they're on the platform but to set up the account and do that first posting and clicking paid links, especially we should have some kind of check in 

Neil: there.

Yeah. I mean, it's, I think there's, there's two different angles to that, right? There's the angle of that's going to impact the bottom line of an ad network. If you have only verified users interact with your ads, make no mistake about it. It's gonna cost some money and try and get that [00:12:00] past shareholders.

Right. That's gonna be difficult. I think the other side on it is people just, I mean, Personally, all of us here are big advocates of, of online privacy of web privacy. Mm-hmm and we do quite a lot of work in that space, outside of our work with, with PPC. I think it does introduce some ethical challenges and questions around, you know, you give your credit card information or you give a passport copy of it to Twitter.

Like, do you trust them? I mean, yeah, I don't, I certainly don't. 

Callum: So I think, think they could just take a charge of like one pence or something off a card, 

Neil: But then what if you do a carding attack with, you know, a hundred thousand stolen credit card you acquired on the dark web. It's, you know, there's ways around that. I think it's a difficult one. I don't think Elon Musk would fix it, but 

Callum: Surely if the name, what I was thinking was if the name matches the card, I know that that's a person that should have different, maybe not a blue check, but maybe a green check or, or whatever it [00:13:00] should have, you know, we could do it so that it has some kind of check.

And I was like, well, we should...

Neil: if you acquired stolen list of card details and, and they're not expensive, right. You know, you're talking 50 cents to $5 per, per card. That's actually active and works in these sorts of places. You're gonna have a first name. You're gonna have a last name. You're gonna have a card.

It is trivial to set up a bot that would set up thousands of fake accounts that then become verified through that. So, yeah, I think it. It's not a losing battle, you know, there's, there's people out there trying to fight it. There's obviously trying to fight it on some of the more advertising side. It's a game of cat and mouse.

I think the ad networks themselves on the social media platforms aren't incentivized to do so, you know, the incentive lies in the wrong place. I don't think regulation wouldn't necessarily help. They'll they'll just sort of move the region that they're based in and try and get around them and you try regulating Facebook and yeah, that's, that's not gonna happen.

Callum: Yeah. Yeah. I mean, I am, I do hope that that something, good [00:14:00] comes of this Elon Musk thing in the way of, you know, just trying to look at things differently at Twitter. Twitter's one of my favorite social media platforms. I do like it. I do enjoy it. . . And I just hope that they can invest in trying to clean up the mess. 

Neil: Yeah. I, I think if they were. And I don't just apply this to Twitter.

I'd apply it to Facebook and such any, any sort of ad network. If they were less focused on the short term profits and the quarter over quarter growth in we, 10% is quarter 15% is quarter I'm more focused on the long term. They would probably realize that if they did take the hit over, say the next four quarters over the next year in verifying their users, they can actually charge more for their ad space.

They can charge more for their interactions because there's a higher likelihood that it's a real person. It's never gonna...

That's why LinkedIn charges more. 

Yeah. I mean, we act, yeah, we, we see, we do see fake traffic on LinkedIn. I haven't said that, but it is significantly lower than what we see on Facebook and what we see on Twitter.

Yeah. I think LinkedIn as well has much higher sort of average order values and [00:15:00] average lead values and so on. So that probably contributes to it. But there's less of a bot problem on LinkedIn than there is on Facebook and Twitter. We actually do see more and more bots month on month coming into LinkedIn.

Perhaps there's a shift happening there, but as it stands, you're absolutely right, and if the advertisers have more trust the integrity in that platform, they're more willing to invest in it. They're gonna pay more for it. And it would work out better in the long run, but short term profits rule everything.

So that's why we are where we are. 

Callum: So, so when we talk about paid media on Google, what sort of say, we're going for a, a Boohoo or a, a general fashion retailer. If they weren't catching those fraudulent clicks, what percentage do you think they'd be getting?

Neil: It's it always sounds like a cop out when I say this, I don't mean it to be, but it's gonna depend on, you know, geographies they're in demographics, they're targeting specific industries and keywords. But if we take a very general look at it on [00:16:00] search traffic around 11% is what we see on average. So to simplify it about one in 10 is where you could put, place it. 

Callum: Wow. So I'm just thinking about some of the companies who have got big, big PPC budgets, like confused.com, the insurance comprise comparison site. I think their PPC budget's about 20 million a year. So 11%...

Neil: It's it's yeah, I mean, we, we actually have, I can't name names, but we have a, a price comparison site in a proof of concept phase with us at the moment and their budget is couple of multiples of that. So you can imagine the, the absolutely crazy amount.

I mean, we work with some massive brands, we work with, Nike, Europecar, Hugo Boss, like big Europcar's not e-commerce, but Nike and Hugo Boss e-commerce brands and, you know, they're spending tens of millions or more a year, huge amounts of money. And when you look at the, not just the amount of money they're losing to invalid activity, which is eye water, you know, the best of times, it's the revenue [00:17:00] opportunity from that.

So, if you imagine 10% of your clicks are invalid, what if you had spent that on real genuine users and they converted at say 10%, right? Yeah. You've lost those sales unless you're running completely unlimited budgets, which very, very, very few people are, you are losing revenue as a result of it. So it's not just a hit, you are getting of, I was defrauded out of.

Hundred thousand pounds. It's that hundred thousand pounds could have generated me 800,000 in revenue. That's what you've lost. So the actually impact is bigger than you might realize at first. 

Callum: So it, it, when it comes to looking at whichever analytics platform you're looking at, you know, say something like Triple Whale or there's North Beam for in Shopify users and the, the popular ones are Triple Whale, North Beam, and I think even, even Google Analytics, you would see that that ROAS number, massively down. Yeah. Due to due to fraudulent clicks. Yeah. So you, you can see that number start changing once [00:18:00] you set PPC, protect up. How long does it take to get set up? 

Neil: Minutes. I mean, is it, yeah, I'm not gonna sit and do the full pitch cuz people no, no truth. But if they want to... 

Callum: I just think it was always, I thought it was always longer than that. I've always...

Neil: No we, so I think we're quite, we're quite fortunate here in the team that we have, we have a really unique team makeup so our tech leads, is he's from the Israeli Intelligence Services, something called unit 8200, which some of your listeners might be familiar with, basically the sort of, yeah elite cyber intelligence unit. Yeah. Within the IDF. He's actually from one of the very elite subunits within that. And I always say if I could pick the five best pupil in the world to build the technologies, we do, I think numbers one, two, and three all work here, so we're pretty covered on that front, but then we're also fortunate where half our founding team come from a performance marketing background. So we're not just taking these innovative technologies. We actually understand the mind of the marketer. And part of that for us was always, marketers are so busy. They're so overworked. They have so many tools and data points constantly thrown at them [00:19:00] every day.

The last thing they want is to have another problem they need to solve. That's gonna take a day or two days or three days to set up. It needs to be simple as possible. So simple tag on the website. Four minutes is what we've measured it out from creation of an account to live data coming through and really focus on that, that simplicity, and kind of running behind the scenes, almost like giving you the data when you need it, but it's just there to do its job and you don't need to have that input.

Callum: Yeah. Yeah. That's interesting. I, I thought it was a lot longer, but maybe I'm thinking about like, You know, the, the traffic coming, through the learning stage to try and see those results. Maybe that's the bit that I'm. 

Neil: Yeah. So we have a learning stage. We have a 14 day proof of concept period. Then in that the system is constantly learning.

So we use a client side, AI function for that. So we actually keep it sort of not too technical. We collect a lot of user information, a lot of models that we run entirely client side so that we are not doing any sort of, what's the best way to put it? Privacy, invasive tracking. Yeah. Everything is very privacy centric.

We've built it to be completely GDPR, [00:20:00] CCPA, and so on compliant. So we don't send any personal information back to our server or back to our systems. We actually do everything on device for the user. But when we update those models, that does take time to build up, you know, a good enough data set for us to say with confidence, this is invalid, this is legitimate.

This we're not quite sure on. So that can take, yeah a couple of days to a couple of weeks to, to fully build up. But that all runs in the background. You don't actually have to have any input on that. 

Callum: Okay. So how are you blocking? Are you blocking by a IP? Are you building out. 

Neil: ...so mixture of IPs and audiences slash cohorts, depending on which platform it is that you're on...

Callum: Right yeah. Cause I was thinking, I mean, yeah, I was thinking about how they do that. Blocking within, within Google talking about, you know, Google. I know we block, I think most people block certain IPs so that, yeah. 

Neil: So in Google we've we've got IP blocking available to us. We've got location exclusion down to the zip code, postal code level, if useful. We've got placement exclusions. So if you're running display campaigns, [00:21:00] lot of dodgy sites out there that you don't necessarily want your ads on. We actually have little plug here, but we actually have a display exclusion list of, I think 65,000 malicious and invalid websites. So can download that from the blog...

Callum: you've you've shared, I use that. So if you're listening, I actually use the exclusion list that Neil was talking about, and it is brilliant. It brought our, you know, for display advertising. Absolutely a godsend I'll remember to put that link for the blog post in the, in the show notes, because you're right that exclusion list that you've got is, really good.

One of the other things...

Neil: we update it every month as well, so yeah, 

Callum: it's so good. It, it was, it is so good. The other thing that I did was, just excluding all the countries, just exclude, you know? Yeah. Like a lot of people don't realize that they should. explicitly exclude, not just go include X, Y, and you know, the UK, Australia, Canada. They need to actually...

Neil: yeah, that's, that's something we've seen, and it's really [00:22:00] strange because there's multiple targeting options for location within Google Ads, so there is target people located in, and then there's one that people often don't realize there are on which is target people located in or who have recently been in, or, yeah, there's one that is people who are located in or show an interest in.

But what we find is when we have people that actually, specifically select the one that's target people located in. So let's say people located in the UK, we're still tracking clicks in our platform that are coming from devices and coming from networks way outside the UK. Yeah. And we're not talking about somebody in the UK using a VPN through the US.

Right. We're talking about data centers, China and Russia. That is very not related whatsoever. And when you actually go in and you manually exclude every country except the UK. And it takes a lot of time, you can use the desktop editor and it's a bit easier, but it takes a lot of time. It actually stops a lot of that activity.

It doesn't stop all of it. Yeah. That's and it's strange, cuz you would think, Google's saying only target people here. Why, why are you getting clicks from Brazil and Venezuela and [00:23:00] all these places? It's strange 

Callum: that that's the thing it's kind of like. They want the fraudulent clicks kind of to go through. Sometimes I feel like, on like I've been caught out by, and I'm sure if you've been in, worked in paid media a few, few years, you've been caught out by audience expansion. 

Neil: Yes, definitely a few times. Yeah. LinkedIn used to be the worst for that, that you, Facebook are very bad too. 

Callum: Oh, oh my gosh. I think I did already audience expansion on REVIEWS.io paid ads and, I didn't have a budget or I had a really, really high budget on this.

I had campaign that I did it on not thinking I only added 10% and I spent like $20,000 in, you know, 48 hours. It was just, and, and they were like, every click bounced within less than a second. Every click was, you know, not in an area that we would be interested in. And they were all, you know, [00:24:00] Android, mobile phones clicking on rubbish, displaying rubbish site.

Yeah. And, and it cost us 20, $20,000 to learn that lesson. And I went to Google and said, Hey, these are, are all fake. And I gave the report and they were said, yeah, we'll give you $150 credit. That's what I got back. 

I'm surprised. You got to be honest. You did better. 

So, oh, oh no. I had to fight for the $150.

They were like, we've approved you for $150. The other 19-20 thousand clicks. That was all legit. And I'm like, how?

Neil: It's ridiculous. And what we find as well is if you look at platforms like let's say Facebook, and LinkedIn's a good example with their audience expansion, you're basically saying, get me more people like these that I'm already getting.

If you are already getting bots and you're already getting invalid users, what you're saying is give me more bots. I want more bots clicking my ads. And that's where we find, you know, something. Our platform and others are [00:25:00] available, you know, it's not just us. Yeah. But something like our platform is super useful because it's actually cleaning that initial data pool that you're pulling from.

So you are saying, get me more people like these real people that are click on my ads, not get me more people like these 10% of bots that click on my ads, which yeah. As you've seen, it becomes very costly very quickly. 

Callum: Oh, it's ridiculous. Ridiculous, maybe is just what annoyed me was the lack of response from the platforms. That was the, the bit that really pissed me off, to be honest. 

Neil: Yeah, we, we have, so we help a lot of our clients with refund requests. So we basically put data in a format for 'em, they can take back to these ad networks and say, we're not happy about this. For whatever reason we want some money back.

Now we always say we can't guarantee anything. We can only give you the data. We'll fight your side, but it, it's kind of out of our control at this point. And we had a little trick that we found, I recommend anyone that is looking to do this, does this. So we uploaded all the data to Google Sheets and Google Sheets has view tracking.

So you can [00:26:00] see who viewed it and when, when they opened it. And what we found was we were submitting these links to Google through their form on their website, where you can fill out, you know, for an invalid click claim or something, they call like. And when they got back to us and said, oh no, actually we're not gonna accept any of these.

We're gonna reject the entire claim. You're not getting any money back. Nobody had viewed that sheet. It had never been opened, no one had interacted with it. So they had no idea what we were actually sending to them. And that's actually quite powerful to go back and say, "Okay, we appreciate your position, however, here's the logs from quite literally your own software, right? Nobody looked at this. Please can you have a look at it and then get back to us?" Geez. And we find the success rate climbs quite significantly from there. And usually you get some sort of bullshit response. It's like, oh, we're sorry, there was an administrative error.

It's it's nonsense. Right. But that's what they'll say, but it does help a lot.

Callum: Yeah, that's, that's a really good tip. That's a really good tip. That's definitely gonna be useful to a lot of listeners. Yeah, I, I just think if they, [00:27:00] how much are they earning off that? How much are they, how much of these networks earning of fraudulent clicks a day is gotta be astronomical, you know, astronomical, you think in the whole globe, the, the millions of you know, accounts paying for paid media, and you know, the novices coming in and going mom and pop shops going right, I'm gonna put my credit card in. Well let's start Google ads. Everyone says it's great. Boom, boom, boom. Yeah. We'll expand our audience. Well, you know, and, and just getting smashed. Because kind of you have to pay to learn almost. 

Neil: Yeah. I would agree. I, I think on the sort of automation expansion, that kind, you know, the, the less control side, I do think that's driven primarily by profit. You know, it's a profit motive. I don't think it's necessarily with the best intention of the end user. I think on the invalid and fraudulent traffic side, I [00:28:00] don't, I don't think it's driven by malice. I don't think that you have a bunch of execs sitting there in a boardroom thinking "Oh great, let's have some more invalid clicks." And you know, we need to make up the shareholder expectations, this quarter. I think it's just a, a lack of caring, you know, as far as they're concerned, it's good enough. Like publicly, they have a good enough face to say. We're addressing this, we're doing something about it. So people will still invest in the platform. If it gets completely out of control, nobody will buy ads on your platform. So you have to do just enough. I just don't think there's enough of an incentive there to really take it seriously. Because ultimately that is gonna hit revenues. And it's like, we're talking about with Twitter before where we say, if you have real users, you can charge more in the long run, but short run. There's some shareholder pain in there that people aren't gonna accept. So I don't think it's malice. I think it's just a lack of care.

Callum: Yeah, definitely. 

And, and, you know, it's a, it's their goal is not aligned with their customer's goal. I think that's the biggest problem that they're not, there's not a shared success really in, in that. Yeah. And it, it sucks. [00:29:00] What's the biggest mistake you see new, you know talking about people coming into the eCommerce industry, you know, maybe setting up their first paid media, program.

What's the biggest mistake you're seeing people make and what should they, you know, if they're listening to this just. Boom. Don't do this. It's probably gonna be audience expansion. Isn't it? 

Neil: That's... I can think of quite a few. Well, one of which would be that. I think two main points I would give, I know you said one, but I'm gonna be a bit cheeky of two.

I think on Google specifically, I really don't like smart campaigns with the exception of smart shopping. Smart shopping's actually quite good. It's moving to performance max later this year. They're quite good campaigns. They can require a bit of tweak and such, but generally they're quite good campaigns, but we see people going and do, like smart campaigns for search ads, smart campaigns for display, and it's just rubbish. So we have one client. We actually a case study from our website Epos Now they're a SaaS company doing point of sale systems [00:30:00] SaaS/hardware. They did a test on smart display, 99.5% invalid activity. No conversions. It was something like, was it $18,000, $20,000 to spend something like that?

It's crazy. So smart campaigns, absolutely crap, don't use them. Other than smart shopping. The second point. And probably the main point that I would say is if you are a small business or you're starting an eCommerce store, or you you're turning over, you know, let's say end up to potentially a couple million a year and you don't have an in-house function to handle this.

Don't do it yourself as a business owner or someone running the company like your specialism. What you're good at is running the business, right? Yeah. Making profit, calculating margins, getting stock and so on. You're not an advertiser. You're not a marketer and it's not as easy as you probably think it is, and it's not as easy as Google and Facebook might make it seem it is. Yeah. So find a good freelancer, find a good agency. There's some rubbish out there, but there's some really, really good ones as well that will help. Yeah. And use them for it. I think far too many people try and do it themselves, and it only [00:31:00] causes problems in the long run. They waste a ton of money and then they come to a conclusion of this doesn't work.

And the reality is it can turn your business from doing a hundred thousand in turnover to 10 million in turnover in a few years, if you get it right. 

Callum: Yeah, no, I agree. I agree. I, you know, there is a lot to be said for bringing in, you know, filling them knowledge gaps with, with outside resources. There is, you know, some amazing, amazing, paid media people out there, but then as we both know, there is some real sharks that yeah. Do it badly. 

Neil: Yeah. I mean, if someone's charging you 99 qui a month to manage your paid media companies run really fast in the opposite direction, don't give them anything. 

Callum: Yeah. I was gonna say that, that that's like the, the Yellow Pages the Yelp.

Yeah. Yeah, they do. They charge that like 299 a month. And you, you are one of how many millions of accounts. Yeah. You know, and you're meant to have a dedicated PPC manager. 

Neil: I would agree. And you, there is, there is software out there that if [00:32:00] you're somewhat software savvy and you do have a bit of time, it can help you if you are doing it yourself.

Think like of Opteo is one that we've used previously. It's fantastic platform and it always highly recommend. So. There's things like that out there that can help, but ultimately getting some professional help was gonna be the best way. Yeah. 

Yeah. No, definitely. I think that's some great advice. I think that's some great advice what I'm gonna do.

I'm gonna put all these details in the show notes so that, you know, people can listen, but also then they can go back and, you know, get these nuggets of information. I'll put all the links that we've spoke about today. Honestly, Neil, thank you so much for being on the show. I really appreciate it. . So I could talk to you all day. I hope to see you in person at, at an event soon, it's been way too long. It's been absolutely too long, buddy. 

Definitely. Hopefully, sometime soon, but no, thanks very much for having me on, yeah, it's been great to speak.

Callum: Cheers, buddy. Thank you.

Callum: Thank you for listening today. In reviews we trust is a bi-weekly podcast where I hope to be bringing you advice and insights from brands that are [00:33:00] taking the e-commerce world by storm.