Saving the Internet

Did you ever wonder who funds the internet? Who pays for all the content and services that we enjoy online for free?

The answer is: advertisers.

Advertisers fund the “free internet”, they are the reason we can enjoy all the content websites (Forbes, CNN, Times), the social networks (Facebook, Twitter, Instagram), the search engines (horizontal search engines such as Google / Bing / Yahoo , and vertical search engines such as IMDB / Pinterest, etc), and much more.

All these services are “free” to us (the consumers) because they are paid for by the advertisers so they can use them as channels to reach audiences, at mass, who might buy the products/services they have to offer.

“If a service is free, you are the product.”

As the internet’s audience grew, advertisers flocked with their budgets and started shifting their spend from traditional media (TV, Radio, billboards, etc) to the internet.

The Promise

Internet advertising came with a promise of being able to reach the exact audience you’re looking for, and actually track the results of all your spend. It was almost too good to be true.

But somewhere, somehow things went terribly wrong:

1. Report: Ad tech (and the garbage #content it funds) is killing the web
2. What We’ll Do When Ad Tech Dies
3. Clickbait And Traffic Laundering: How Ad Tech Is Destroying The Web
4. Google : End Of The Online Advertising Bubble
5. Marketers thought the Web would allow perfectly targeted ads. Hasn’t worked out that way.

…and here we are.

I will not get into how we reached the current state of digital advertising, The articles I listed above go into great lengths to discuss this. But I want to talk about something else.

Simply put, internet users hate ads. They’re intrusive, suck up their precious bandwidth / limited mobile resources, get in the way of them benefiting from the websites they like, and offer them no additional value. So what do they do?

Block the ads, install ad-blockers, fight ads, call for an ad-free internet, etc.

But then what happens to the publishers who depend on advertising revenues to fund their websites and keep the lights on?

Well they can either start to charge customers (subscription fees, pay-walls, etc), or shut-down their websites. This doesn’t work well for the readers of those websites, who by blocking the ads on the websites inadvertently lose access to this free content when the website runs out of money and shuts down.

Here is the vicious circle:

1. Publishers produce content / services for free
2. They make money through ads
3. Readers enjoy these services for free
4. Readers don’t like ads
5. Readers fight ads and block them
6. Publishers don’t make enough money to keep their websites running
7. Publishers shut down
8. Readers lose access to the content / services they were enjoying

This is a zero sum game for the readers, publishers, and advertisers.

How do we solve this?

Well we need to understand why do people “hate” ads. And I am not just talking about internet ads, I am talking about ads in general, including the commercials you see when you’re watching your favorite TV show (and which you most probably Tivo through, Tivo is TV’s ad-blocker)

I believe people hate ads when they’re not relevant.

By definition, advertisements are “messages paid for by those who send them and are intended to inform or influence people who receive them.”

In its purest form, advertising is a form of information. But information is only valuable to its relevant audience. If you send information to someone who is completely not interested (non-relevant audience), suddenly it goes from being a valuable and informative message to annoying spam.

Online users are spammed left and right with advertisements that are irrelevant to them. Which turned the internet into a never stopping TV commercial.

The advertising industry tried to solve this problem by using “behavioral targeting”, and an entire industry grew around trying to profile online users and selling these profiles to advertisers so they can figure out what they would be most interested in.

Take the re-targeting industry for an example, you visit an e-commerce site and check out a bag, and then for the next 30 days almost every page you visit online has that same bag showing in a banner somewhere. It keeps following you across the internet, almost like a digital beggar who won’t stop following you until you give them what they want.

Users are profiled relentlessly by tens of vendors across any website you visit. All you need is to install Ghostery to your browser to see the amount of trackers that are harvesting your info in hopes of selling to advertisers who can then “target” you with more relevant messages.

Unfortunately (and unsurprisingly) this still did not solve the problem. Just because the advertiser got some information about a user’s profile, and assuming it is 100% correct (which is not always the case), this still does not mean that the user wants to be blasted by the “relevant” advertisement everywhere they go across the internet.

So what’s the solution?

Enter Semantics

I believe the answer lies in the second school of advertising, namely “Semantic Advertising”

Semantic advertising is all about targeting ads based on the meaning of the content a user is reading. It’s that simple... or is it?

This was the initial focus of AdSense when it started. But the problem with AdSense is that it’s not really semantic, it’s just contextual.

Let me elaborate on the different schools of advertising a bit.

Behavioral advertising says “put this ad for a car on this page even thought it is about sports because a few clicks ago this user was looking at a site with car reviews”. There are many successful ventures doing this, in fact this is the majority of all advertising being done on the internet today.

Contextual advertising says “put this ad for a car on this page because the page mentions cars”. That is an multi-billion annual business for AdSense (gross simplification alert). Think of contextual as a poor man’s version of semantic.

Semantic advertising goes one step further, it should be capable of identifying the context and (more importantly) the meaning of the words on the page, in order to determine the appropriate advertising. This is an advancement to the current contextual model where the simple identification of keywords is seen as sufficient to represent the context of the entire page.

Why can’t AdSense (for example) try to be more semantic? Well because to be truly semantic, you need deep “vertical” knowledge in the domain you’re targeting. Are you doing semantic advertising for healthcare, financial products, retail, etc.. ? These are all very different verticals and need very different knowledge graphs. And Google can’t possibly do that for AdSense across every single knowledge domain on earth, so AdSense remains “horizontal” across all domains rather than “vertical” across one domain, and it does contextual (not semantic) advertising.

So the majority of advertising on the internet today is (1) behaviorally targeted (which is considered a huge improvement over no-targeting), followed by (2) some contextual advertising (the biggest player being AdSense), and then (3) barely any semantic advertising at all (a quick visit to this Wikipedia page will show you what I mean.

But why?

Mainly due to two reasons:

(1) Ad-tech companies built incredible hype and sold advertisers on the fact that behavioral advertising is the “holy grail” they’ve been missing all their lives, is the most effective way to advertise, and that you’re missing out if you’re not doing it.

What this actually meant is that once any advertiser has managed to build a rough profile about you, they can blast you with their ads wherever you are on the internet. But you should be ok with that because the ads match your behavior, right? Wrong. Internet users seem to disagree, and the rise of ad-blockers tells that story pretty well.

(2) Doing proper contextual advertising is more difficult than behavioral, and doing true semantic advertising is significantly more difficult than either.

Plus there is no hype around that nor are advertisers asking for it, so why bother or care?

Well, we need to start caring …

Because if we don’t, we’re fighting a losing battle against internet users, it’s just a matter of time. Users hate digital advertising in its current shape and form. And unless this industry undergoes a major re-haul, it will simply crash and burn, and it will take down all the companies depending on its revenues down with it (publishers being the first). This isn’t good for the internet nor its users.

I see many companies, experts, and authors talk about how to fix this. But they all seem to be fighting the wrong battle. They keep talking about things like “improving viewability”… Improving viewability of something nobody likes?

In my opinion, there is only one way to fix this: Make ads relevant and beautiful. Get them to actually add value to the content rather than take away from it. If users get value from the ads, they won’t fight them, they will actually engage and see them as something valuable that is integral to the content they’re consuming.

The Shoppable Web

A couple of years ago we asked ourselves a simple question:

What is the purpose of an ad?

The answer took us back to the definition of advertising: to influence someone to eventually make a purchasing decision related a product or a service. Period.

All the advertisers care about at the end of the day is to get consumers to buy their products or services, they need to sell, and to sell they need more eyeballs, and to get more eyeballs they need to advertise. And then that takes the customer on a journey that ends up in a purchase somewhere down the line.

This is a fairly accurate representation of the digital customer journey [1]:

The distance between “awareness” and “purchase” is quite significant. And obviously customers dropout through this funnel and a small percentage eventually move to each subsequent stage.

So we had an idea: what if we shortened this journey considerably and combined all the steps between awareness and purchase into one single step?

If we managed to allow users to buy a product at the moment of their peak interest, wouldn’t that help increase conversions and just be great? We certainly thought so…but where would that point of peak interest be?

We believed it is when the customer is reading content related to the product or service, that’s when they’re the most engaged. If they’re taking time out of their busy internet lives to read content about something, they must be interested in it, otherwise their fleeting attention would be directed elsewhere almost immediately.

AdRelated is born

So we set out to create a product that can understand any piece of content, and recommend products or services that are semantically related to that content. We kept playing around with this idea until we eventually created what is now known as AdRelated [2].

AdRelated is the world’s first semantic product recommendation engine, it is a technology that can make any content “shoppable”.

The way it works is pretty simple: a single-line javascript widget that any publisher can install on their website. The widget then automatically analyzes the content of every page it is loaded on and shows readers products or services that are related to the content on that page.

Let’s saying you’re reading an article about the Samsung Gear VR, AdRelated will show you the product being sold on e-commerce stores in your country, Google AdSense will show you…well see for yourself:

AdRelated vs AdSense

What about if the content isn’t talking about a specific product or brand? No problem, AdRelated will understand the topic and still show relevant products or services. If the article is talking about the last soccer match between Real Madrid and Barcelona, it will show you the Nike shoes Cristiano Ronaldo was wearing in that match, or the jersey Messi was wearing. If you’re reading about travel to Madrid, the widget will show you hotel rooms in Madrid, or airline tickets to Madrid, you get the point.

The idea is to always show the customer something (mainly a product or service) that is highly and semantically relevant to what they’re reading. This way, we send the right message, to the right customer, at the right time.

→ Right Message + Right Customer + Right Time = Beautiful Ad

AdRelated in Action

 

Saving the Internet

We believe that all the content that exists online can be much more effectively monetized, and rather than serving as a dumb billboard for ugly banners that readers hate, it can actually show ads that delight and bring value to people.

AdRelated can be a cornerstone in building what we call as “The Shoppable Web”: all the web’s content can become shoppable, and every page can be a mini-store.

Instead of customers searching for the products, the products are now searching for the customers, and we shorten the customer journey and shift it from a theme of search to that of discovery.

Discovery is what Pinterest, Instagram, and others did. But instead of building a destination website of our own, we can transform the whole web to be that destination.

Instead of fighting a losing battle with internet users trying to shove ugly and irrelevant ads down their throats, we can make ads beautiful and useful. We can get users to like and engage with ads, and help publishers to monetize their content in a sustainable fashion.

This will not just help save those publishers, but in the long turn I believe it will help save the internet as we know it.

If you’re a publisher or an advertiser then I invite you to join us, and help us save the internet.

— Moustafa

—?—?—?—?—?—?—?—?—?—?—?—?—?—?—?—?—?—?—

[1] Organizing for the Digital Customer Journey,
https://www.b2bmarketingexperiences.com/2015/08/organizing-for-the-digital-customer-journey/

[2] AdRelated makes content shoppable and just might save ad-tech,
https://geektime.com/2016/01/19/adrelated-makes-content-shoppable-and-just-might-save-adtech/