Tag Archive for 'Duncan Riley'

Online advertising – a bubble

I just recorded a podcast with Duncan Riley and Bronwen Clune – two New Media innovators I greatly admire, to discuss what the future of media was. Unfortunately, the podcast recording came out battered and my normal analytical mind wasn’t in gear to add fruitfully to the discussion.

So Dunc and Bron, here I go: why I think advertising on the Internet has a future that will repeat the property bubble that fueled the world’s economic expansion these last few decades. (Y’know -the one that just burst.)

Advertising has been broken by the Internet
Let’s think about this from a big picture first: why do people advertise? It’s to get an outcome. Ignoring elections and government campaigns, the regular market economy has advertising so companies can make money. Pure and simple. Whether it be "brand" advertising which is a way of shaping perceptions for future sales, or straight-off-the-bat advertising pushing a product – the incentive for companies is to get a response. That response, ultimately, is to take that cash out of your wallet.

Now’s lets jump into the time machine and think about companies in the 1970s and 1980s – before this "Internet" thing became mainstream. How could companies get exposure for their products? Through the media of course. The mass media had captured audiences, and they were able to monetise this powerful position they had in society by forcing people to consume advertising as they were dealt with servings of information they actually wanted.

It worked in the past, because that’s how the world worked. That is of course, until the Internet and the Web completely transformed our world.

Companies jumped on the web thinking this was simply an extension of the mass media but so much better. And they were right to some extent – it was much better. A bit too good actually, because it now exposed the weaknesses of the concept of advertising.

Take for example one of the undergraduate students that works at my firm. Apparently, this 19 year old never watches television – but he is on top of all the main shows. He does this through peer to peer technology, where he is able to download his favourite shows. I asked him why does he do that and he responded quickly: "because I can avoid the ads". What’s happening with the Internet is that consumers can control the experience they have when consuming information now, unlike the past where they marched in line according to the programming schedule. The audience is no longer captive.

The Internet did another thing: it made advertising more accountable. In the past, savvy agencies would ‘segment’ the population and associate various mass media outlets as better being able to connect with the ‘target market’. To measure, print used circulation and readership – working out how many people bought the publication, and some number out of some Actuary’s head of how many people read that same copy (through statistical techniques of assessing patients in doctors’ surgeries, no doubt). Broadcasters on the other hand, would randomly call households and using statistical methods, would estimate the number of people that tuned in.

Perhaps the fact I took statistics for my undergraduate degree, is why I am so skeptical. Even my stats lecturer admitted it was bullshit – albeit in an ‘educated’ way. In relation to the mass media, the bigger issue was the fact this educated bullshit was not disaggregated. What I mean, was that when a newspaper has a readership of 100,000 people – there is a massive assumption that if you advertise in that publication, you will actually reach them. You might have bought a newspaper to read this one article your friend mentioned – and yet, your act of purchase enables the newspaper to justify all the other pages to advertisers with a simplistic metric.

The Internet completely changed this because we no longer are relying on statistics, but actual data collected. In the past, advertisers would get a plane and fly over an Amazonian forest they picked and pay to drop one million pamphlets hoping that at least 50,000 of their target market would catch the pamphlets and respond. Of course, indirect sales activity could indicate the effectiveness of a campaign, but in reality it was all a guess. Now with the Internet, a lot of the guesswork is not required any more – and quite frankly, advertising on the Net looks bad but the reality is that the truth has now been set free.

This is looking at it from an accountability point of view, but looking at it from a practical view as well, there are issues. The holy grail of advertising, is targeting. The reason being, if you can target an ad better, you are more likely to get a conversion. However there is a natural friction with targeted advertising and it’s called privacy. As I’ve said before, privacy is the speed hump for the attention economy.

Advertising on the net technologically offers a great ability to target, with marketers licking their lips at the opportunity. However this is coming with a complete misunderstanding, that technology may be an enabler but culture and society will be a breaker. People do not want better targeting. The thought that some company profiles you scares the crap out of people. Yes, I’ve even convinced myself that when advertising is relevant, it’s useful – but this is looking at it after the fact. The problem with targeted advertising, is that whilst it may run a world record 100 metre dash, it might not get the chance to actually get off the starting blocks. Just ask Facebook if you don’t believe me.

The structural impact the Internet has had to ruin advertising
The Internet is great for measuring – but there are a few too many measures. The lack of a consistent measurement system creates several problems. More significant is the fact that different types of Internet services compete based on what model works best for them. For example, pay per action is something advertisers love because they are getting a better return on their investment by seeing a follow through. This works with contextual advertising like the kind Google uses – it’s actually in Google’s interest for you to click off their pages.

Contrast that with video sites where a person is engaged with the content for ten minutes. An advertiser can’t compare ten minutes of engagement on a video site easily with click-actions on contextual advertising sites. What this creates is a vacuum, where the ad dollars will bias those that offer a better likelihood of making a sale. After all, why would you care about capturing someone’s attention for ten minutes, when you can simply pay for someone clicking on a link which is directly linked with an e-commerce sale on your site.

This creates a real problem, because it’s not an equal playing field to compete for the advertising. Certain types of services do better under different models. Banner advertising will die, not just because people are realising the usability issues surrounding banner blindness , or the fact that banner advertising is simply a copy and paste model of the mass media days , but because competing advertising models that better link them better to final sales will become more popular. When we hear about the great growth rates in online advertising, don’t forget to dig a little deeper because the real growth comes from search advertising which makes up about half of that.

There’s another structural problem with the Internet: there’s too much competition. In the mass media days, the media had an established relationship as "the" information distribution outlets of society. With the Internet, anyone can create a blog and become their own publisher. Additionally, the Internet is seeing growth not just in New Media ventures, but utility and commerce ventures as well. Same advertising pie theoretically (ignoring the long tail effect for a second, where small advertisers can now participate), but a lot more "distributors". This creates a fragmentation, where advertising dollars are being worn thin. It’s for this reason the larger internet services tend to manage to get by . Just looking at the face of it though, you know there’s a problem in the longer term even for the bigger players when you operate in such an environment.

It’s not just other Internet services to worry about however: it’s the advertisers themselves. In a world of information, democratised by search engines judging quality content – you as a publisher are on the same foot as the company paying for the ads. Why would Nike want to advertise on your website, when it can just improve its own search engine ranking? Companies can now create a more direct relationship with their customers and future customers – and they no longer need an intermediary (like the media) to facilitate that relationship. That’s a Big Deal. It’s not just search though – the VRM Project is doing exactly that, creating a system that will facilitate those relationships.

Concluding thoughts
I could just as much put an argument in favour of online advertising, don’t get me wrong – there will be a lot of growth occuring still. But what I want to highlight, is that taking a step back at the facts, there is something seriously wrong with this model. If advertisers no longer need that intermediary to facilitate a relationship; if advertisers are chasing the industry down the tail of measureable ads that better link to a final sale; if the entire industry is not consistent and competing with each other both in inventory and in methods, in an infinite battle; and if consumers are no longer captive to the content distribution experience – it makes you question doesn’t it?

According to Nielsen over a year ago, about a third of all U.S. online advertising dollars spent in July came from the financial sector–with mortgage and credit reporting firms representing five of the top ten advertisers. Together, those companies spent nearly $200 million on search, display and other Web advertising, meaning that a slowdown would degrade fairly significant annual revenue streams. The writing was on the wall that long ago, what analysts are only now saying are troubled times for online advertising.

Just like we knew a year ago about the credit crunch, before a drastic turn of events turned it into the most dramatic economic shift in our world in our collective memories, so too will the advertising bubble burst. It will be years – perhaps decades – before this happens. However one thing is for sure – the Internet has not only ruined the newspaper, music and traditional software industries, but it’s also ruining the world of advertising. Like how newspapers, music and software are currently evolving into new models which we still are not sure where they will end up, so too will advertising be transformed.

Mr Online Advertising and Ms Media Company relying on it as a revenue model – you are growing on the basis of some very shaky foundations.

Organisations need to be a size 12

Last week at the top 100 web applications launch, Ross Dawson made a brief remark that I feel should not go ignored. He said that technology aside, companies like the ones in the top 100 list have a huge impact on our society; They are redefining our society as a whole, with new ways of doing things such as how organisations are structured. On that same panel Duncan Riley was crying out foul about the problem with Australia is that venture capital money is nothing like how it is in the US (which is offered for riskier ideas, at a quicker turnaround, and with bigger amounts) – but a retort made by Phil Morle brought this common whine in the Australian industry to a different level: “You don’t need to be a billion dollar company to be successful.”

In the context of the discussion, this can be taken in several ways about the state of the venture capital industry and its interaction with Internet start-ups, but take a step out of that mindset and instead explore the opportunity with the point made by Ross. Organisations, like incorporated companies of today, are something we should and can re-examine because a billion dollar company is not what the goal should be. Why? I’ll show you.

(Dis)economies of scale

Economic theory proclaims that the larger a company, the better. In the literature, this is regarded as ‘economies of scale ‘ whereby things become cheaper the bigger firms become. For example, the cost of capital is cheaper for a large a company (for those without a financial background, that simply means things like the interest on a bank loan or how much a shareholder expects as a return in dividends or share-price growth) . Operating costs can also fall, which if you think about how retailers will offer discounts on bulk buys – if you are a bigger company, you buy more and therefore get those deals (as well as have influence to create those deals).

Even if you don’t have an economics background, I am sure you are familiar with the concept given the ‘growth’ obsession we have in our world: bigger is better or more is more. However something we should be equally aware of is the ugly cousin: diseconomies of scale . Even economic theory recognises that you can get to a stage where you are just too big, where in fact each extra increase is no longer creating economies but the opposite. It’s a bit like trying to carry the shopping from your car boot: some people can carry a half dozen bags to save on multiple trips; however there is a point where they are carrying too many bags, and the extra benefit of less trips back to the car is in fact outweighed by the increased risk of dropping the bags and hurting their back.

We live in a world, where the growth obsession of our world fails to recognise the ugly cousin. We constantly hear about growth, but what goes up must come down – we never seem to hear when a company is “big enough”. Building on Ross’s point, maybe the answer to that is not that we need to identify the point on the continuum where the diseconomies kick in; instead, the new opportunities offered by technology can instead determine how we can organise resources with the least amount of size.

I have a client that is regarded as one of the biggest advertising agencies in the world. I’m sure anyone with experience with the internal operation of ad agencies will recall how damn complicated they can be – which I think has to do with the ego prominence of the creative industries. Everyone needs differentiation in those industries (and so, the one company is in fact a group of multiple agencies like their own mini empire or stand alone business unit). The complex organisational structure that my client operates in, made me think this is what modern day socialism is like: create a large organisation that becomes so complex, that no one understands it – and in the process, have multiple over-lapping jobs filling functions that are not needed. Giving people jobs for the sake of it. As a case in point, one of the guys in the finance department told me how there was a girl that no one knew what she did. One day she resigned, and whilst one would expect strain on a group with one less staff member, what actually happened was that no one noticed any difference in the output of the team!

Little did I realise however that soon afterwards as I performed a internal (non-client facing) role that this advertising agency wasn’t unique with its socialism. Aside from the fact I’ve met people at my firm that I still couldn’t tell you what they do, my experiences had me see another bigger negative about a big organisation that can be summed up in one word: people. And just like how people by nature are complicated, so too will my answer as to why.

My firm employs 140,000+ globally and about 5,000 in Australia. Whilst that is a high number, more remarkable is the fact it’s a professional services firm: we are not talking about 140,000 high school drop outs but a well educated work-force. As a consulting firm, client facing staff like myself can be in a group as big as 200 people (in each city). There are effectively another dozen or so people with exactly the same skill set and job role as me, but we are just resources that go out to different clients, so ultimately we are doing the same work. That side of my job at my firm has seen me experience a very efficient, lean machine with the fundamental economic concept of “allocating scarce resources” brutally evident with the language of how we run our projects (I even just called myself a resource above, not a employee). However it’s that internal role which had me see the supporting ecosystem for client-facing groups like mine and which made me realise the weaknesses of a big company. I couldn’t tell you how many of the 140,000 people are supporting the client-facing professionals, but I would hazard a rough guess to be about 20%. The nice way of saying it, is that in Porter’s model , that 20% are the support staff to help execute our primary revenue-generating activities. Another way of putting it: that 20% are the overhead.

Overhead matters for two reasons in this discussion: it slows down an organisation (ie, decisions) and it can definie its existence (ie, costs). Each of these points are worth looking at separately.

lego men

“Frustration” defined: the sum of all people you need to work with to get something done in the last month

People and decisions

That internal role I discussed above, was about implementing a new technology at my firm. I could write a book about the experience, but suffice to say I can recall one incident which is a perfect reflection of something I learned about getting something done in a big company.

This particular technology allows you to add extensions that can drastically alter the functionality of the product. These ‘plugins’ are remarkably simple – we are talking about uploading a single file that is perhaps 50kb (smaller than a typical word document) – and once uploaded via an admin interface it can be activated for immediate effect (with documentation fully provided on the web). Indeed, in the early days of the technology’s roll out, I would often add new plugins as I felt the need arose, but that quickly ended when I was forced to concede that’s not the right way to do things (as it’s not my job). So therefore, if I ever wanted a plugin, I would have to e-mail the IT guys, who would then review it, test it, and then upload it. This is fair enough because adding a plugin could destabilise the system losing valuable data.

I might also add whenever I sent those e-mails after I no longer installed those plugins myself, I would follow up a few weeks later only to find no one had got around to doing it.

Why the significance of this story? Something that I could do in 30 seconds instead takes weeks because I work in a large organisation. For example:

– write an e-mail asking for the request explaining why: 30 minutes

– following up on the status of my request: 60 minutes of e-mails, listening to justifications for inaction, etc

– escalating to a superior when I felt things were taking too long: 60 minutes of meetings and e-mails, as I stressed the importance of a particular plugin for the productivity of one of our pilot groups

…And that’s in raw effort. That’s not accounting for the time stretch of a few weeks (actualy months in one of the cases).

Even though I had the skills, understanding, access, and ability to do this – I couldn’t due to lines in the sand of what I was allowed to do. And because I couldn’t, what would take 30 seconds for a small start-up using the same product, it would in fact take me hours upon weeks to get another few people whose “job” it is to do it.

This example is more humourous than harmful, but when it comes to large organisations, it’s a perfect characterisation of how things get done. I can assure you, all big companies work like this – by definition, a company that is Sarbanes Oxley compliant has so many segregation of duties it will make you cry with laughter. If I shared with you some other stories, that laughter will turn to shock, when you come to the truth of how companies actually operate.

People and cost

At another one of my clients, they have been undergoing some massive growth over the last few years, with a large organisational re-shuffling as this rapidly growing company took shape. A guy I’ve got to know that’s been there a while (and which I might add, we have no idea what’s he actually does as a job despite his title) told me something quite funny. His observations over the years, is that as a company increases in size, so does a corresponding increase in headcount irrespective of any other factor. By example, he explained that lets say a new person is appointed to lead a new team – they now require a personal assistant. And that new team now needs a dedicated IT guy. And then an HR representative. And the list goes on – rather then a company consolidating on its size (ie, merging job roles to avoid duplicity), what he thinks is that as the company has grown over the years, there is always a corresponding increase in head count regardless without any obvious reason why. It’s almost like a natural externality of growth is headcount.

Payroll is a significant cost for any company which can be up to 80% of the total expense of an organisation (my former headmaster told me that, that being a knowledge intensive organisation: a school). So as my above discussion highlights, I obviously find it amazing how such a significant cost is not controlled because management don’t actually understand what staff they have (and as an aside, current enterprise social networking technologies specifically target the real need of documenting what expertise a company’s staff actually has because no one knows). Whilst this may seem like an important point from a controlling costs point of view, I wish to raise it’s actually a hell of a lot more significant.

Let’s say a company needs one million dollars a week to pay for things like wages, electricity, office rent etc. In other words, a company needs sales of minimum $1million a week purely to stay alive – to pay for the stuff that in theory is meant to help it make money in the first place. This is without regard to meeting profit targets as expected by a company’s owners and other such factors. If that company can’t cover that $1million, it is technically insolvent. Meaning, jail time for the directors and senior management for running such a company.

So if a company has these commitments, it ties their hand. They suddenly become very risk-averse; where experimenting with a better way of doing something may threaten their ability of making that $1 million a week. Couple that with the fact that most organisation’s single biggest expense is payroll, with lists of employees that no one person exactly knows what they are doing, and it makes you wonder. Companies effectively exist to cover their expenses, but if they actually dug down, they’d realise those expenses may not even be something that require. In effect, a company’s entire strategy and positioning in the market (ie, prices that take into account enough to cover overhead) may be dictated by something that might not be needed. A big company exists purely to feed the beast, making decisions that may not be what a company should be making if it didn’t have to worry about its overhead.

small is the new big

Small is the new big

The power of 12

Going back to how technology is enabling us new ways of organising, if the only reason why a company needs to get big so it can get economies of scale, why don’t we flip it? We no longer live in the industrial era where economies of scale are the goal. Instead, the biggest cost we have now is time; if the expertise is in the people we employ, we need to scale operations so as to give them money and working conditions that suit their lives. This isn’t relevant only for professional services, but for any web service – the fact you provide a service and not a manufactured good means its driven by people not metals.

Of course, a company’s strategy can either be cost-competitive (like Dell) or differentiation (like Apple). But this doesn’t negate the fact, that a company should only have costs that directly add value for the customer (which is why we have innovations like activity-based costings which allocate overhead directly to customer activities, but that’s another story). Given that people are the biggest expense in companies now, we need to question, do we really need that many people to provide that value?

I have a rule I follow in life which has grown out of my experience with how things go wrong: complexity. The more factors, the more likely something is going to fail. For example, if you are driving to a wedding – the more traffic lights, the more likely it is going to slow you down. The longer you have to drive, the more likely you will get involved in a car accident. If you need to drop something off on the way to the wedding venue, if you need to make two separate stop-offs, the more likely you will be late as opposed to one drop off (regardless of the distance, but based purely on unforeseen variables). So basically, you need to minimise the ‘nodes’ in the line. Even when things look they are fine, the more variables to success, the increased risk of something happening that will distract that goal.

However, I am willing to concede, you can’t be a minimalist for everything, which is why I am settling for the number 12 – keep your variables, especially people, to a maximum of 12. Aside from religious connotations which makes the number so omnipresent in our lives, the reason I like it is for the same reason it’s the base number of measurement systems used through history , like the still dominant imperial measurement system. That reason being, it’s one of the most versatile numbers. Versatility and agility to the market is what success is now; not economies of scale.

If designing an organisation, you want a core team of 12 people. Those 12 people together, should be able to do everything that needs to be done to meet the needs of the customer. And if the organisation needs to scale for whatever reason, then those 12 people should have specific functions that they own. The number 12 is magical, because for the same reason it was used in the measurement systems in the past, it is so versatile: you can re-group your 12 people into even teams of two, three, four or six. Don’t underestimate the impact team dynamics like that can have – or using a term that Mick Liubinskas says as often as a nun does her Hail Mary: it emphasises the importance of “focus”, in an agile way that can easily adapt to situations. Yet at the same time, you will find with 12 you can get almost anything done if you truly have a talented team. Still don’t get 12?

The number twelve, a highly composite , is the smallest number with four non-trivial factors (2, 3, 4, 6), and the smallest to include as factors all four numbers (1 to 4) within the subitizing range. As a result of this increased factorability of the radix and its divisibility by a wide range of the most elemental numbers (whereas ten has only two non-trivial factors: 2 and 5, with neither 3 nor 4), duodecimal representations fit more easily than decimal ones into many common patterns

Source: Wikipedia

There are dozens and dozens (whoops, was that 12?) of companies that are small but successful . Challenge yourself: does world domination really equate to a 15,000 person workforce? Focus on getting 12 highly capable people, and you will avoid entering the trap of the big companies today that are slaves to their own existence. We have technology today that could design a radically different organisation in 2010 completely foreign to how traditional business operates. If you explore what smart people have said to complement what Ross originally meant by his comment, you now might also realise that those top 100 web applications represent more than meets the eye.