We're stating the bloody obvious. It's a digital marketing podcast, for Christ's sake. But I think where I was going was just with everything that you do, all the comms that you're doing. Try and think about where that fits in in that lifestyle. Is it the education, the persuasion, the credibility, the capability certifications that you mentioned before? You know, where is it, what you're producing? Where is it? How does it need to be delivered?
Yeah. I mean, speaking to clients yesterday, and they said exactly the same thing. Like, you know, when a buyer is talking to them, they're sort of eighty percent mind made up. Yeah, they've done all their research. They know what they need. They're just needing that extra bit of like pricing information, whatever. But they're just about there already because they've consumed enough information to make that decision already. You have to be aware of that. You've got to be providing information for the right people at the right time.
What?
Is it working?
Um, yeah, I think so. Welcome back to Digital Marketing From The Coalface. And, um, I will need help and guidance today.
There's a very, very blank sheet of paper in front of you.
I've had a yeah, I've had just had one of them weeks at the coalface and I have been away to a wedding as well. So yeah, a midweek wedding. Yeah.
Was there enough people? Did people take time off to go?
There's people there from Australia.
I mean if you're going to go from Australia, it doesn't matter if it's a Saturday, or a Wednesday.
It's great. It was a an amazing venue. Good mate of mine, Graham, who sometimes listens to the podcast. Sometimes.
So you have to say it was a great wedding.
It was his, um, youngest daughter. And yeah, it was good. It was good. Um, week's a bit upside down and I've been chasing my tail a bit in the office yesterday and today, but hey, it's the weekend, so who cares?
Yeah. Um. Oh well, I have a couple of things which
It's fortuitous.
could inspire a healthy debate or could take five minutes. So it could be a long one or a short one. I don't know which way this is going to go, but basically it's to do with the the statistics that ninety two percent of B2B buyers enter formal evaluation with a vendor already in mind. And what does that mean for.
Where did you get that from?
Um, that is a Forrester report from 2024. Um, so what does that mean for websites? The sort of general idea of what I was looking at was that, um, if you're like hyper focusing on CRO for B2B website.
Conversion rate optimisation.
That's not really the- you're kind of missing the point because they've probably come to your website just to go: "are they who they say they are? Do I feel comfortable? Yes, I'll get in touch." So they've done a lot of thinking already. The other statistic there was like, there's often six to ten decision makers involved in making that decision. And some of them will have been on your website. Some of them won't. They'll have gotten information from other places or being told by someone else. So whoever's placing the order is almost certainly going to come on and get in touch. And it's almost, you know, just don't scare them off. Don't obsess about if the buttons in the right place. And you know, if the form's pretty, but just make sure that there's nothing on your website that is going to stop them finishing what they started. Um, so I thought it was quite interesting. It's like.
I mean, are they basically saying that like most businesses, referral business.
Doesn't have to be referral business.
How do they find out about you then? And then decide they're going to use you?
I think it was saying that a lot of it is referral business. But they might have seen something somewhere else. They might have been to a trade show. They might have seen LinkedIn.
If they've only done it from a trade show perspective, then what persuaded them that you're likely the people that they want to use?
Yeah.
If it's not a referral, then it must be the website and all the material you've got out there. Yeah. That persuaded them that you're the people that they want to use when they reach out.
It could be the website information feeding into the decision making process earlier on and somebody else has looked at the website, it could be that they-
But what was the statistic again? What was it actually saying?
Ninety two percent of B2B buyers enter formal evaluation with a vendor in mind. So you could probably interpret that in many ways.
Yeah. I mean, to me, I'm reading that as they've done the research. Like you said, they've come in via AI engines. They've come in via Google, they've come in via paid search, LinkedIn, social media, wherever they found you. They've consumed the information they've thought: "these guys seem to know what they're talking about," and therefore you're on the list. So when they actually then reach out to say, can you give us a price for or can you give us a proposal to do time kind of thing? Is that what it means? Because it is a bit confusing.
It is a bit confusing. Um, it also says forty one percent have one preferred vendor before you start evaluating.
And again, that could still mean that they've come to the website, they've done all that.
They might have done.
Before they've even spoken to you and thought, yeah, these guys, they've done it for these guys. The case study, this is, like, what we want them to do for us. And therefore, unless when they actually engage with you, you come across as an idiot, and then suddenly they go, oops.
Yeah. Or it could be that the engineering people have been on your website and got all the information, downloaded the data sheets or whatever, and then go on to the procurement person and said like, yeah, go and get a price. Yeah. And the procurement person goes on and and goes, oh, wait a minute, they don't have the certifications. Oh, well, this doesn't look right. I don't think we should be using them. And their criteria are very different from what the engineering people would be. Yeah. You know, they do this and the tolerances are this and it's very, very up our street. But the procurement people might be looking for different sort of validation. So, you know, who are you? Who are you designing for?
I interpret that as like the website is vitally important because it's doing the job of persuading them that when they shortlist that you're on that shortlist, they're not coming to you stone cold. Like, can you do this? They decided you can.
The information on the website has somehow persuaded them. It's more about the don't obsess about how you're converting them once they're on there, because that's not how they're behaving. They're reading stuff. They're going away. That's okay. Because that's, um, part of the process. And then the person that comes to actually fill in the form doesn't really need to be pushed down a path of converting, because the reason that person is on your site is to fill in the form.
I mean, CRO is just as important in that situation as well, isn't it, in the persuasion stage? It's not just about making sure the buttons are the right colours when they click it. It's about telling the right story, making sure the information is easy to access, making sure the social proofs are there, all that kind of stuff.
What they were saying was, um, it's much broader than that. It's about the social proof. It's about the case studies. And that's probably the most important thing is having reassurance. Um, you can do the job. You've got evidence, you can do the job, you've got all the right, you know, certificates and things so that, you know, you can do it well. They were saying that procurement is pretty much like loss aversion, risk avoidance. Nobody's going to go like, oh, this website's really modern and really interactive. And therefore I'll press the buttons. It's like this website looks like an engineering website should look and it's got all the things it should have, and it's telling me the things I need to know to make me feel safe in terms of filling in this form. And when it's getting to that point, it's very much about making them feel safe.
I think it is at every stage. I think at every stage, you know, the way that you present yourself to the world is all about making people feel like if they decided to engage with you, they're not going to look silly. They're not going to have awkward questions to answer when, you know, a lot of money's disappeared, um, into a project that didn't deliver what it was supposed to deliver. And then somebody says, well, why on earth did you ask them in the first place? Any idiot could see that, you know, they weren't going to deliver. They don't look credible.
They've never done anything in our industry.
Yeah, they're not big enough. Yeah.
They're not a member of this body or whatever. And it's very much like that.
But in today's world, you know, we do like a car analogy. Most people will do a load of research and then decide what car's right for them. Very few people... I mean, I noticed on AutoTrader, there's a thing on there where they're asking you, you know, about your lifestyle and what you need a vehicle for and all, you know, all that kind of stuff to try and then guide you. I've never used it. But I mean, that's, you know, I can see the how that might be useful to some people. But for me personally, when I'm looking at cars, I do so much research, it's ridiculous. And all I need then- And Hitesh will be the same.
I'm married to one of them. I mean, honestly, I was looking for- when he was looking for a car. It took about three years to buy a new car because he was researching it in the minutest of detail, and he spent all his spare time watching car videos. Yeah, it was terribly tedious.
It is. But the point is that then went to find the car with the spec that he wanted and an organisation that would sell him it with a warranty and they were credible and they weren't like, you know, I mean- not too long ago, Caron and I went, um, to look at some cars and, you know, they immediately... we even said like, we gave them a budget and they immediately showed us something that was five grand over our budget. And it's like.
You're not listening.
No, you're not listening to me. I know it's their job. They think they're clever salespeople. They came in, they wanted to spend twenty five and they bought this car off me for thirty because I showed them how the finance works and all the rest of it. No, you know, that's not how I operate. It's like. We know we can buy what we need for this budget. Have you got one? Yes or no?
Yeah. That's, um. That's what my husband did. He, um he went in and said, can you get me this specific car in this price range, this old, you know, can you find me one? And they found one that he wanted.
Mhm.
But yeah.
But it is about, as you said, I mean, we've gone over this plenty of times in the podcast. You know, I think people get really confused in their comms and they don't, you know, you have to look at everything you do from your buyer's perspective.
Yeah, exactly.
From your customer's perspective. What's making them tick, what's keeping them awake at night? What do they need to read, watch, or consume in any way that makes them feel like you are likely to be able to make their life better? And that sounds really dramatic, but if you're a procurement person and you've been tasked with finding suppliers of X, then you just want to make sure that when the company signs the purchase order or signs the check or sends the money to the bank account or whatever, you know that you look good.
They're going to deliver.
Yeah, they're going to deliver.
Yeah. Yeah, absolutely. Um, the thing you were saying about like doing the car research, that was one of the other stats. Seventy to eighty percent of purchase journeys are complete before the buyer talks to anyone. So you're nearly, uh.
This has been the case for the last ten years. Yeah. This has been a mantra which people have gone over and over and over again. Like people are doing tons of research and like with the AI engines like Perplexity, ChatGPT and Claude, the level of research that you can get into is, you know, you will almost, you know, like for a long time, people have gone to the doctors and told the doctor what's wrong with them. And, you know, all right, some people are going to get that spectacularly wrong, but quite often they get it right. Yeah. Because they've done a lot of research. They're often better informed sometimes than the people they're talking to.
I went to the doctors years ago, um, with a really weird thing going on my foot and um, this was a specialist at a private hospital, specialist in whatever I was going for. And he looked it up on Wikipedia.
Yeah.
Because he had no idea.
I had exactly the same. I had a thing called Stevens Johnson Syndrome, which is kind of potentially it's, um, ten percent of people who get it die. I mean, it's serious. Yeah. And yeah, like four GPs had looked at it and they didn't have a clue. Then another GP looked at it who was about to retire Doc Taylor and he went, mm, I've never seen this. And he was about to retire. I've never seen it but I think it's Steven Johnson syndrome. It's a reaction to penicillin, or not penicillin. It's a reaction to a trimethoprim or something like that. And it's a weird one. It's a syndrome. So they can't do anything about it. And if it gets worse, it's horrible and it can be bad. Or... It was bad enough what it was doing to me, but, you know, my body dealt with it. But he then said, right, go to the specialist. First thing she did was look on Google because, you know, why would they know all about it?
They can't know everything.
They might see it twice in their career. Yeah, yeah, I don't know how we got on to that. But anyway. But the idea is, you know, that people are doing tons of research and as a result of that, um, yeah, I mean- what that introduces obviously is the idea that when they're doing tons of research, that's when you also want them to find you. You know.
You've got to be in front of them.
Again, for stating the bloody obvious. It's a digital marketing podcast for Christ's sake. But I think where I was going was, you know, like just with everything that you do, all the comms that you're doing, just try and think about where that fits in, in that lifecycle. Is it the education, the persuasion, the credibility, the capability, the certifications that you mentioned before. You know, where is it, what you're producing? Where is it in that? Or how does it need to be delivered?
Yeah, I mean, speaking to a client yesterday and they said exactly the same thing. Like, you know, when a buyer is talking to them, they're sort of eighty percent mind made up. They've done all their research. They know what they need. They're just needing that extra bit of like pricing information or whatever. But they're, you know, they're just about there already. And because they've consumed enough information and found out enough to make that decision already. So you have to be aware of that. You know, you've got to be providing information that's for the right people at the right time, which means you need all sorts of, you know, it's not just one person. You need all sorts of different bits of information.
Yeah, it's like, who are your tribe? Where do they hang out? What sort of information do they consume? Again, you know, anyone that's followed the podcast will know that we've been banging this drum for the last three or four years we've been doing it. Four or five years, however long we've been doing it, but it still seems to miss. It still seems that there's plenty of people out there, you know, that don't get that, they don't understand that they need to be presenting things from the customer's perspective.
Yeah.
What are you looking for?
Um.
More stats?
Yeah.
Oh, yeah. We need more stats! Goodie! More numbers.
Yeah. That was the thing. If it is a referral, that's somebody spoken to someone else. If there's six to ten people involved in the, um decision making process, there's still between five and nine people who haven't had that conversation, who haven't been convinced by the conversation at trade show or in the pub or whatever. So you still have to work out where they fit in. Somebody's going, yeah, this is great. And they're going, yeah, is it though? They might have a different angle on it. They might actually not be that convinced. So they, you know, they're needing persuading even though someone else is.
Yeah, I mean, the referral thing is hugely important, isn't it? I mean, we probably, as an organisation, we should probably ask for more referrals because we know we probably don't mind that to rich seam as well as we could.
We have occasionally and it never got us anywhere.
No. I mean the most recent referral that came in, I mean yeah, they kind of turned into tire kickers. I mean it hasn't gone anywhere at all despite, you know, going down to the south of England to go and see them and getting really good vibes and having a really good meeting. And then pffft! They just went quiet. So much so that like the last time I tried to reach out to them, they didn't even bother replying. And it's like, okay.
It's very odd, isn't it?
Yeah. Yeah. Maybe they just didn't like me. That's fair enough. Plenty of people don't. Why should they? I'm horrible.
Oh, okay. Um, yeah. So yeah, I think it's-
You could have disagreed, but never mind.
Ah, nah.
Um, yeah. Visibility is as important as actual conversion. Getting in front of them and, you know, being where they're looking and, and not obsessing over, you know, the color of the buttons, sort of thing, which in B2C people can get very, very obsessed on like, you know, because you've got big numbers. It's like, well, why did only fifty percent of people click? You might have like sixty percent of people who've clicked. It's huge numbers. And, you know, with B2B, it's just such a very different situation. And, you read stuff about, you know, conversions and stuff and it's easy to go, oh, that's the reason, you know, and it probably isn't.
I can see, you know, certainly like looking at on page conversion optimisation, um, very much in the realms of B2C, you know, it can absolutely transform a business in a B2C environment. The way that what you say, how you say it, the layout of the pages, colors of buttons, the number of fields in a form, etc. it's very, very different in B2B. The people in B2B are arguably more motivated. It's a bigger problem they're trying to solve. There's more money involved. It's, yeah, it's complicated.
Yeah. They've put a lot more thought into it before they get that far.
I mean, a rational argument.
You know, a recent, really nice enquiry, which we're currently working on, it's an accountancy business consultancy organisation down near Coventry. Down in the Midlands.
I like working with accountants.
I used to do a lot of work with accountants.
Um we have worked with accountants in the past. Yeah. And they were very clear that they came to us because of our HubSpot expertise that was important to them because they use the HubSpot stack, but they're not using the CMS content hub.
The actual
website.
And so yeah, they found us, I think it was an organic visit not a paid search visit.
No.
From memory. But it was very much, you know, so you've got that expertise. So. Right. Okay. So then they land and then we managed to persuade them. So they reached out, filled the form in. But then I think in our line of business, most agencies are the same. Like, that's hard to get that enquiry. But once you've got it, um, it's then all about that first conversation and trying to understand and putting a bit of effort into again, trying to understand where they really are because, you know, being able to build a website with HubSpot is fine. There's plenty of other people who can do that.
Yeah.
You've still got to then further differentiate and make them feel safe.
Yeah, and they have to work closely with you. So they have to think, yeah, you know- they have to like you. Because, if you're buying a pair of shoes on the Internet, it doesn't matter if you like the person who's putting them in a box or not. But if you're working with people and building your website, you've got to have that sort of connection or rapport or whatever you want to call it. So it's, um, it's not even once they filled the form and you're not all the way there.
Mhm. Oh no. For sure.
I mean, if, yeah, if you're buying a piece of kit, maybe it's slightly different, but for a service business like ours, it-
Especially, you know, identical kit. Yeah. It really is just price based stuff. That's it. Because it's the same.
Price, delivery time or warranties, whatever it is. Yeah. Mhm.
Okay. So yeah, I mean, I guess it's an interesting stat and it's actually kind of shows how stats can mean lots of different things. On the face of it, it's like, you could interpret it as kind of it doesn't matter about your website really because you know, they've already decided kind of thing.
But how have they already decided it probably is still down to your website. It's just maybe not what you think it is.
That's right.
And you've got to look at more of it. Um, another thing I was thinking about, and we've been in this situation a couple of times with clients. So say you've got an ad campaign running, you've got your-
What type of ad campaign, just-
Say Google ads, for example, or LinkedIn ads or something. So you've got ads, you've got a landing page. Um, people are clicking on the ads. They're not converting.
Mhm.
At what point do you decide: is it the ad? Is it the landing page or is it a shit product? You know? And we've had clients come to us saying we need you to get us more conversions. And we've throw more money at it. We've changed the landing page. We've changed the ads. We've changed everything. And there's just no more conversions. It's like, you've used as much- That's all there is. There's nobody else wanting what you're selling. You can throw as much money at this as you like, but the market is a finite size and that's all there is. And we can't get more conversions because there's nobody in the pool willing or even thinking of converting at that point, but it's quite hard sometimes clients are, you know, you're like, no, no, you just have to get us more conversions!
I agree, and I think- I'm trying to remember who it was. I mean, I know we don't talk about specific clients on the podcast, but there was something recently where they were very specific about the number of enquiries that they wanted and the business they were trying to generate in terms of numbers. Oh yeah. I do remember who it was now. Um, and when we were having a meeting with them and I said, you know, have you actually done the market research and established that...
The size of the market.
the business is out there. Has the market been saturated? I didn't actually say has the market been saturated, but like what they sell is very it's very specialised, you know, like how many opportunities are there?
Yeah.
In that part of the world?
Yeah, exactly.
Not a UK based company. So how many opportunities are there like, um. Uh well, yeah, didn't know, but I mean, there is, there is data out there. I mean, you may remember, um, the SaaS tool that we built Alertise, you know, separate business to Red Evolution, and it's still being used, you know, by, uh, mainly by companies in Australia because that's where Steve's based. He was the brain behind it. Um, but that was one of the things that in strategy execution and even strategy planning is to try and understand that.
Yeah, what's the size of the market. And how much can you realistically get out of that?
That's right. I mean, you might look at it and go like, you know, over the last two years, you know, the amount of business generated by this product line or service line has been reducing, there's something wrong with what we're doing. But if you look at the bigger picture, go and look at the ONS data. It's actually like you're bucking the trend. There's plenty of businesses that have either disappeared who were doing that or, you know, they're doing way worse than you are. You're actually doing okay.
Well, that's it.
What is the baseline? The benchmark?
Yeah. The market isn't as big. Or is there another competitor that's come in, or has something shifted? Some disrupter come in and, and people just don't buy the thing in the same way anymore. And just putting ads out there and expecting to reach a certain number isn't always realistic.
Marketing, I think is misunderstood. And, you know, and, I'm obviously heavily involved in marketing, but unlike you, I have no formal marketing qualifications. And I guess over the years, my view on marketing has changed, especially, you know, running this business. But I guess on the one hand, a lot of people like me would think of marketing as, right, the clever people have created this product. Mhm. Go and sell it.
Yeah. And that sales.
That sales, that's not marketing. Marketing is we're in this area of business. What is this area of business crying out for? Well, actually, you know, like, let's say, for example, um, we're a firm of accountants just to stay on theme with accountants. And we're trying to understand, you know, what are the pain points in the marketplace? And it's like, well, you know, customers find, um, for example, um, keeping a log of their expenses an absolute pain. Let's go back a few years before all those tools exist. You think, right, okay, so if we created a tool that was like, you just take a photograph of a receipt and then it just gets added into the accounting system, that might have a market. Yeah. Okay. Like so you then.
Then go and make the product.
Then go and make the product, because you've established the market exists and that's the role that marketing played. The market research.
Yeah. That's it. I mean, we've had this conversation before-
But it's dead easy in the current environment to do market research. To a certain extent, because you can go and look at search trends. You can go and look at search data. You can go and look at the data for the traffic coming to your website or the people's websites, the generic tools like SEMrush and all that. And you can very quickly establish like, oh, there's huge demand for this.
Yeah, yeah. Are people looking online for this thing. Um, and yeah, and then, you know, who are they? You know, and, you know, if you're selling baby products, you've got to look at the number of babies being born. Is that going up or down and things like that. Yeah. The information is all out there.
Mhm.
If you're in the property business, there's all the trends of like, you know, how many houses are being sold and what the interest rates are doing and things like that. So yeah, the information is all there. There's no excuse really for trying to come up with unrealistic numbers and advertise more. Because that'll get more in. If people aren't there wanting what you do. You know, it's the, um, build it and they will come thing isn't it.
That's right. You know, I mean, if you think about um, thinking about campaigns and you mentioned Google ads. So if you think in B2B anyway, a lot of Google Ads campaigns start out as basically fishing exercises, and that's with an F, not a p h.
Yeah, we're not trying to scam you.
We're not trying to scam you. It starts off as like, like we don't really know. The market's got, you know, it's maybe not saturated, but there's plenty of people in the marketplace providing software tools that do what this software tool does that we're trying to market. So you're going fishing. You're trying to understand what the trends are, what people's pain points are. And it's no more than that.
What's gonna hit, you know.
Yeah. And I think that's why a lot of people-
You use it as research.
It is research. And a lot of people don't get that with paid search. They think that paid search should be, you know, every click is like, why hasn't that turned into a, you know, a sale? Why hasn't that turned into an enquiry? And a lot of the time, it's, you know, you're trying to actually figure out.
Is there a market there?
Yeah.
Or if there is a market, do people look on Google. Do people search online for this thing or are they tied into somebody they've always used and they don't bother? So at which point, there's no point. Don't throw your money at it.
We've got a new new customer just now. And they're in the software space in a very specific niche area of software. And they're trying to get market penetration over in the United States. And they don't really know. They don't know whether they can actually compete with the established players. They don't know what the opportunities are. They've carved out a few niches. They've got customers over there. And I think they got it that this is like initially anyway, it's-
It's just a test.
A test. It's a test. It's an investigation. It's research. Research that might have- that has rather that definitely has the potential to generate business. And I think that's how I often communicate.
A lot of the time, it's a lot cheaper to chuck something out there and see what happens than conduct formal research, which is really, really expensive. You stick something out there and at least you're learning about the ad. So if it does work, you've got information there and you've got data and you've potentially even got some sales. And if it hasn't worked, it's still cheaper than doing a personalised bit of research.
Yeah. And that was often the conversation we had back in the day where people were, I mean, people still are kind of like skeptical about paid search about Google ads and like, because a lot of people threw money at it. Use the tool badly and just, you know, just-
Spent a lot of money.
Yeah, exactly.
Nothing worked.
But back in the day when, you know, people say, oh no, I don't want to do any paid search, I just want to go for SEO. I want to optimise, I want people to find us organically in Google and that's fine. But if it takes you six or seven months to actually get established in organic search, only to find out that people don't want to buy what you're offering.
Yeah.
And you could have done that in a couple of months for a lot less money using paid search. It wouldn't have felt good to be spending money on Google clicks and getting nothing from it, but you would very quickly find out that you were barking up the wrong tree.
Yeah. Nobody actually wants this or nobody looks for it. Sometimes it's like, oh, they actually don't call it that. You could optimise loads of pages on your website for a specific thing. And then you find out your customers call it something completely different. So you could spend a lot of time, you know, putting a lot of content together around something and it just doesn't hit, whereas if you do ads, you could say, actually nobody's, there's no volume on that. But look, there's people searching for this instead. And it really helps your content creation process.
Mhm. Yeah. Okay.
What else have you got?
What else have I got?
Anything?
Got a couple of little bits and pieces. Oh, Google is finally going to be introducing AI impressions separately in Search Console.
Alright.
It's rolling that out in the UK first.
Okay.
So some people in the UK have it already. So, um.
Is it going to be giving you phrases that appeared in AI?
Yeah, I think so.
Like it does with-
Yeah. So it's only impressions. It's not counting clicks, which yeah. Reasonable. Because yeah, a lot of them aren't clicks. Yeah.
Okay.
But it's coming. There's some...
But that'll just be Google properties. Gemini?
Yeah. It'll be AI, AI overviews and AI mode, I think. Um, I'm not sure. It's called generative engines. So I think it would be Gemini as well.
Generative engine optimisation as they call it. GEO.
Yeah. Something like that. But yeah, some people have got it. We don't, I haven't seen it in any of our accounts. I had a look.
Even if you spoof being somewhere else, you know, like you use a VPN and say, yes, I am in America, give it to me as well.
No, they're rolling it out in the UK first because of some legal thing that the UK government's um, given Google... had a word with Google. So it's actually rolling it out in the UK first. But no, I've had a look through some of our client accounts and our account, and haven't seen it yet, but I have seen pictures of people in the UK-
Where did you pick that information up?
That's from one of the search engine Journalers. Yeah, one of them. Yeah. So it is happening. So hopefully we'll see it soon.
It's just something else for people to worry about isn't it? Fret about.
It is. But I think it's interesting.
GEO, instead of SEO.
To see you know, where it's going. And you know, if it's important to, you know, be a lot more informed as to like, whether it's important or not. And at the moment we're going, well, you better look at this. Most of it's probably coming from Google, but you better look at this. But to be able to see, well, actually, you know, you've had like one hundred impressions from search and two hundred from Gemini or AI mode or the other way around. I think it will help with content creation. It'll be more than just sort of guessing that you should do a bit of both. We'll be able to see what the priorities are.
Yeah. I mean, it really is just an extension of SEO in most respects.
Yeah, yeah, it's a variation, but, um.
There's little things you can do which improve the way that the tools can consume your content. But yeah, they're very much aligned. The two things.
Yeah. If you're looking for data in terms of like what's the most volume in terms of what people are searching for, it's giving you a lot more information on what people are actually searching and how they're searching. So it makes it easier to do it right, which is good.
I think it's worth pointing out, you know, you're not going to be able to bypass SEO and just do GEO instead. It's like they both require you to have credible presence online, credible content, answering questions, all the stuff that you haven't been doing for SEO.
Yeah.
You know, don't think like, oh, I'm glad I didn't do all that.
Yeah. It's just a variation on what you were doing before. It's just a slight tweak.
Yeah, I mean, a lot of the content that we've got that gets consumed by the artificial intelligence engines and regurgitated as citations, etc. it's content that we've been producing, you know, for years. Yeah. I mean, all right, we are playing around with how we deliver content. In order to try and get more citations and get more opinions.
A lot more granular.
Mhm.
You know, instead of generic stuff. It's um, but yeah, I think people are asking much more specifics.
That trend was happening in normal search anyway.
Yeah.
It was. Yeah, I remember, I don't know if it was Eric Schmidt or somebody years ago who was talking on Google and he was saying, you know, the average length of a query has gone from two words to seven words or something like that.
People were already being a lot more conversational with it. But yeah, even more so now. So it's just a continuation of what was happening. But yeah, I think it'd be really interesting to get that data rather than just guessing.
Um, yeah.
That's also, um, I sent you that tool. There was a, um, content grader thing and it, um, graded your content in terms of was it commodity or non commodity, which was really interesting. Like could a AI LLM engine thingy, um, machine.
Stop talking technical!
It's Friday, I'm hungry. I'm like my brain's now struggling with all this, but yeah, could a machine have written it or how specific is it to you? How much of it could have been written by machine, how much hadn't like things like, you know, have you mentioned, you know, "when I was doing this thing, I noticed this specific thing" and "ninety percent of our customers said," and...
Yeah.
So I thought it was quite good. So you can run a piece of content through it. I think it was like Claude or-
How different is it to the machines? The tools that have been around for a while, let's say, whether something was definitely or possibly or wasn't written by AI, is it just a variation on that theme by the sound of it?
It gave you a lot more information. It was like, this bit is garbage and this bit is better. And it gave you recommendations on how to make it better, but it was quite good. So that was, uh, it was some sort of like skill that somebody had written. I think it was Perplexity, but there's various graders around. But I thought- that was one that really struck me as being quite practical and giving like hints as to what to do to make it-
I mean all this-
It's weird because the engines, they want specific, and they're writing garbage. So it's like, if you want really good stuff, why don't you write really good stuff?
Yeah, because it can't.
It's weird.
Yeah, well, I mean, it can up to a point. I mean, it kind of smacks to me a little bit like, um, you know, the old saying that, you know, to get on in life to succeed. What you need to do is to master, you know, be sincere, you know, and once you learn to fake that, you've got it made. You know, it's a bit like that. You know, it's kind of like, yeah, I don't know. And how much does it really matter? You could look at a piece of content, read some content and go, right, this is what I needed to know. And I've read this content and I now know that. And it didn't offend you. I mean, most people don't know good writing from bad writing.
No. And don't care.
And I mean, and don't care. I mean, you know, there are wildly successful authors. And if you actually give that to an English professor at Oxford and ask them for an opinion on it, they would throw it in the bin and say, this is absolute garbage. Well, it's a million best seller, people love it.
I've read books that have like no paragraphs or just like, or no chapters or they're like, no punctuation. Yeah, there's things in there, like hugely successful, because-
So, so we keep, you know, we're kind of very keen to kind of point out just how awful AI content is, but like we've consumed awful human written content.
We have.
For millennia. You know, I mean, it's like.
It's not some- I think this was not so much, you know, it's awful content. It was, um, weirdly, the AI engines like stuff that's very specific.
Yeah. Well they're like conversational stuff as well because they're basically with the LLM, they're trying to understand how language works because the way that they get better is by understanding the relationship between words and how language works. So when they analyse a bunch of words, the way that they do it, which I did read about how they do it, and I was kind of none the wiser at the end of it. But anyway, you know, when they analysed that, the better, the more granular the model, the more conversational content they've consumed, etc., the better they get at mimicking that conversational content.
I've just remembered What it was trying to get at, actually. Told you, my brain's a bit slow this morning. It was saying that, um, if your content is commodity, it'll just get quoted in the overview or whatever. It'll just get regurgitated. If it's very specific and non-commodity, it'll have to link to you, right? Because it can't answer the question sufficiently. So it then sends people to your content. Okay. Because it's got all the examples and they can't sort of just summarise it and it's not bland enough for them. So that's what it was. It was more like if it's non-commodity, they have to send people to you. And it was about doing that so it was useful in terms of like, if you had a bit more personality or add some like real life case studies into this, then you're more likely to get actual people looking at it rather than it just getting sort of paraphrased.
Okay.
So it was a good tool.
I mean, to what extent are people primarily using AI as a chatbot? Do you think most people are still using AI as a chatbot?
Lots of people aren't. I mean, I think like DuckDuckGo has got more users now because some people want to use things that don't have AI.
Okay.
I mean it's...
But what I mean is, you know, I think commonly people perceive AI as a chatbot, as a way of like, you ask it questions, it gives you answers as opposed to it doing stuff for you.
Yeah, yeah, mostly. And most people are just using it to ask, you know, yeah. I don't know what's the weather going to be like, I don't know.
Because, you know, we very much use it to help with production. We use it to do stuff. I mean obviously-
Take these data, summarise the obvious. Put it in a table.
The obvious one is coding. You know, Phil uses it extensively for writing codes and is highly efficient as a coder because of that and doesn't seem, you know, fazed by it. He's been you know, he's been kind of consumed by it.
It's not putting him out of a job. It's just making him ridiculously fast.
Yeah. I mean, I, you know, a simple example, like yesterday I was closing a competition. I think I mentioned this to you and like as part of closing a competition, a golf competition, you have to make sure that everybody has used a different marker. If there's three people playing golf and two of them put down the same person as the marker, then in theory you should be disqualifying them. But the way that the software, which is pretty clunky, um, has been written is you kind of have to go in and open a drop down menu and look at the person's individual scores for each hole, make sure the signatures are in there and see who the marker was. And then in your brain log that and then go through the other sixty five and make sure that name doesn't come up again. And every time you look at another name, you've got to think have I seen John Smith's name before?
I'm sure it was John Jones. Yeah.
So I just got called to do it for me. Claude just goes and looks at the detail of the web page, the document object model, the Dom, and analyses all the data and says, oh, this name appears twice as a marker. So he goes straight to it. And, I think that's a very simple example, not work related leisure related of using AI to, to make my life simpler, because the software doesn't do that one thing it should do, which is like, show duplicate markers and it doesn't do it.
It's not difficult.
Not difficult, no.
Or anything. It's just a bit of a pain.
And I think once people wake up to that, like, you know what I would say to anyone who's like skeptical about AI and they think it's just a chat engine is just think of anything that you need to do that's really tedious and think, can AI do this? And is it worth me spending ten minutes to show the AI what to do and how to do it? I mean, we're lucky. I mean, we use Claude and Claude has the cowork facility, which is like running locally on your machine, on your files, and it can use the Chrome extension to pretty much do anything. So anything that you can do in a web page, you can tell Claude to do it instead.
Yeah.
And analyse stuff and grab information.
I'm still getting to grips with that, but I'm trying to use it more.
It's surprisingly slow at times. But it's incredibly powerful. Plus you can just tell it to go off and do it while you do something else.
Yeah, I find that, yes, I actually like, I kind of have two things I need to be doing at once. So I get it to do something while I do the other one. And that was quite handy, except it didn't really do it properly. It had a bit of a meltdown. Claude was really having a bad day yesterday. Everything I asked it to do. I was like, can you use this text and make it into like a summary paragraph. It was garbage. Try again. No, that's actually worse. I'll just do it myself and then and come up with some straplines. That's terrible. Come up with five more straplines. No, they're also awful. Again, I just had to do it myself. So sometimes the human brain...
I had it helping me compose something like it was, I think it was an email or something. And it did. It was just dreadful. And that's the good thing. You look at it and go, you know, that's dreadful. I'm not going to use it. I guess the challenge is for people that go, well, A's done it great. I'm not even going to read it. I'm just going to copy and paste it and send it.
You've got to be.
You've got to be careful. But after I've done it and I actually said to, you know, to the engine, no, this is garbage, I'll just do it myself. And its response was, well, when you've done it, can you share it with me so that I can try and understand why what I did was so awful, just like a person would really, it's like gave up. And so I did. I fed it back in and, and then what it pointed out to me was bang on that. It's like, yeah, you know, I was really verbose about this bit and you were very concise. And then I did this and this and this and that. That wasn't even needed. You didn't even put that stuff in. You know, I'm a bit terse with emails anyway, the best of times, but I did think it was quite funny.
The thing it did yesterday was um, I was getting it to draft something and um, it came up with this. You know, the number of enquiries in this field of work is increasing. There's no reference there because I've taught it that every time it comes up with the fact it actually has to reference where it got it from. Like, where do you get that information from? And it went, yeah. Well, there is no evidence. I'm sort of guessing that this means this and therefore this means this. But you know, you now have three options. We reword it to make it more vague. We find some facts to back it up or we do something else. I'm like, yeah, well, go find some facts to back it up. And then it went, I can't.
Mhm.
Like.
Just like people do. Yeah. People make shit up all the time just to bolster their case.
I mean, it might be true. You could- all the other facts make you think that that could be happening. But there was absolutely no evidence. And this sort of stuff we're creating here, it actually needed to have evidence. It wasn't the sort of thing you could just make rubbish up. So yeah.
That's the thing. And, you know, in many walks of life, you know, that's where AI is dangerous because if it makes shit up, then yeah, that's I mean, you know, they've all got the disclaimer at the bottom.
This is AI, it can make mistakes.
Some of this, you know, might be absolute garbage. It might be wrong. Yeah. Well, yeah, you can understand why in some walks of life that's a bit of a challenge.
Yeah, exactly. I mean, what I was doing was sort of like the referring to like legal stuff. So I mean, you can't just make stuff up there.
No, no. Yeah. All right. Um, anything else particularly you want to talk about?
No. I've written all these things and I'm bored with it already, so. No.
Okay. All right. So you've been listening to, uh, Digital Marketing From The Coalface with Dave and Julie.
Have a nice weekend. It won't be the weekend when they listen to this, but you have a nice weekend.
Hope you've had a nice weekend as we're hoping to. Lots of golf this weekend. Looking forward to that. All right.
Bye.