Hasn't this been done before? If not, I can't believe it. It strikes me as a great idea. And there is no reason to limit it to print advertising, especially since print and media is so often combined in ad campaigns.

I think you could sell advertising-related advertising (designers, artists, marketers, webtechers, printers, whatever). And maybe sell advertising at cost per-hit.

Great luck to you.
Wes

something like http://www.hatads.org.uk/ ?
Tommy

@landsky - it hasn't been done to my knowledge. it's surprising to me too because it seems like a "lexis nexus" research type tool like this amongst advertisers would be very popular.

@Tommy - that appears to be more of a "museum-like" physical collection. they claim to have pieces online but I was unable to access it. I'm less interested in preserving old ads for posterity and more interested in extraction of practical lessons from seeing how recent ads have evolved.

My friend Ed suggested an interesting mutation of this idea: aggregate multiple versions of online display ads run by businesses and track those over time. You get into issues of questionable practices and violation of TOS's when you're accessing peoples ads with an automated scraper and inflating their impression numbers and it seems like it would be harder to extract lessons via the evolution of the ads since most are rotated automatically by the adserving mechanism. But interesting nonetheless.

sean

I like this very much, imagine being able to revisit different decades like the 80s through ads. Even through ads associated with magazines. Seems very clever way of looking at ads in general.

vanhees exactly! Although, i could easily see how his could be differentiated than that one. That one is just a collection of ads, his could be an evolution of a product through the viewing of its ads i assume.

Clever Idea!

Cathleen

i like it

"80s through ads" There are actually several printed books in marketing that show in-detail the ad styles and individual elements used. For the decades.

You can get them a Barns and Noble and also at Borders Books. Look under the graphics design section.

Check out this 80's ad:
http://www.youtube.c.../watch?v=8uyK8JXP5dM

There are a few websites that have ad examples posted.
http://abduzeedo.com...mate-80s-inspiration

Anyways, I think it would be a nice idea to highlight elements of design online. Sounds like a viable idea to me.

@kevin - "revisit the 80's through ads" has nothing to do with the idea I proposed. That was introduced by Vancouverbluz in a comment. I have no interest in creating a site that let's people reminisce about the 80's - i'm sure there are plenty of multimedia resources out there on youtube already for that. I'm interested in the service proposed in the idea subscription and want to make sure this comment thread doesn't derail the discussion into something completely random, useless and unrelated to the original idea.

sean

Well, I never assumed it was just the 80's. I thought it was for all time decades was it not?
"A web-based research tool for people in advertising that allows them to see how print ads have evolved over time. "

@Kevin- sorry if the time scale wasn't clear from that description. it's really more intended to show iterations of individual ads over a period on the order of months. The "evolution of advertising since the 80's" concept has more of a novelty "Smithsonian-type" feel to it rather than something which could be used to extract practical marketing lessons.

Ultimately I just want access to the lessons and collective wisdom of all marketers over the ages who have fiddled with running different versions of their ads.

sean

the cigarette companies have done this. im in the ad industry and i dont see this as a viable pay service. the capital made is from new ideas. there are plenty of books that chronicle historical design, mags or otherwise, that to me are more valuable resources. however, that is only one person's take. while im not sure of the profit angle of subscriptions, perhaps something more in the direction of what google is building with scanning in libraries full of books, for free. and then of course they profit from the traffic, perhaps that model could work. i would definitely dig through it as a resource if it was free.

I think it would make more sense to split of the design elements into sections and time. That is where the style is from, you can tell ad styles by the colors and the way they advertise.

There are certain templates that are designed like this. For example take a look at infomercials. Notice how they all sell to the user in the same way. But, wait there is more you will get a free gift. All for only 3 easy payments for $19.99. But, if you call in the next ten minuets you will get a limited time special deal and only have to pay 2 payments of $19.99. So, act fast and call now!

@jakrose - can you point me to examples done by the cigarette companies? This idea seems to be constantly misinterpreted as a "history book" collection of advertising, which it is not. Think more like an Apple Time Machine interface for stepping through iterations of the same ad in a magazine to see how it's altered. If you're in the ad industry and you're aware of a service like this, please post the URL so I can check it out. I wouldn't nix the viability of this as a subscription model - it's something I want myself right now.

@Kevin - I just don't understand your last comment and the informercial reference. The ads would be related to their corresponding ads and tagged with time info so they could be stepped through over time.

sean

You know like elements of the ad. You are trying to tag things like
toon illustrated, ads that use clouds in them, ads that are all black and white.

good idea!

Have you considered applying an insight filter (service) to this idea? Whether it has been done before is irrelevant. Anyone could track the changes over time, but WHY the changes were made would probably be more relevant to end users. Since no one knows exactly why the change was made, maybe you create a community for people to hypothesize the insight driving the change.

You could also leverage a private network where people are familiar with specifically that category.

Most graphics designers will notice why ads used certain elements. Even most people should be able to pick out the elements. I don't really think you need to tag every little thing. If its the main concept and implemented well, you should be able no notice.

Have you ever wondered why they have SUV's on a rocky landscape, with mud over the hood? When clearly they are advertising for city drivers that will never encounter such an event. I think they made there concept pretty clear. It like gee, I why they put that girl with a bikini by that sports car.

Advertisements are usually pretty clear with the concept.

@Steve- brilliant. that's actually the first truly valuable suggestion made on this thread. having a service that makes it easy to see changes over time would give folks a reason to congregate but the real value and staying power as you suggest would be in the social component of allowing others to share theories around why the changes were actually made. No reason to make this a "hub-and-spoke" interaction that isolates users from one another. _awesome_ idea.

@Kevin - by tagging I mean identifying the ads across multiple issues that are of the same original version. Keep the service simple- it's just a way to see how ads change over time. Incorporate the social element steve suggests and allow people to easily converse and overlay their theories as to why the changes were made and you suddenly have a very interesting service that becomes a community no ad exec can live without.

sean

This was an interesting bit:
http://www.lifeismar...7_07_01_archive.html

love this idea! It would be awesome for either getting a great retro type ad or for the forward thinkers, "future" ads. I like it Alot

I like this in the respect that you are applying data mining ideology to obtaining the ideas from past ads. I know that marketing is somehow related to culture. Parhaps you are making projections from what has been done in the past. To what could be done in the future. Interesting.

With respect to technique its mostly the same thing over and over. There has only been a few changes to it over the years. Mostly there are style changes rather then technique changes. I would focus on those.

I like the idea. Will I be simply shown the evolution of the adverts on one easy-to-use page to save me the bother of searching, or will you do some of the analysis for me and highlight trends? The former, while perhaps less useful, also seems much more straightforward. The latter sounds more like what you're proposing, but how do you plan to make it work?

I'm not in advertising and will readily admit I know little about it - would an analysis that simply picks out devices on a yes/no basis really be useful? For any more sophisticated trends than that, the analysis will only be as good as the person doing it. Can you make enough money to pay someone good enough?

@PhillipH - i'm proposing to provide just the infrastructure that allows people to scan, categorize and upload ads from consecutive magazine issues. As the first contributor I would seed it initially with a few tech zines so there's something of interest in there for people to talk about. The commentary would be provided by people viewing the ads and proposing hypotheses about the factors at work and why the changes were made (see @Steve_TI's suggestion above).

Revenue model is either a "freemium" subscription (if the service is useful enough to bear it) or ad-driven. In either case, contributors of categorized ads are rewarded with a fraction of the revenue proportionate to how much those ads specifically generated. In the ad-driven scenario, this is straightforward and merely a royalty on ad sales. In the subscription scenario, it would be a function of engagement generated by their ads and therefore how much their contributions drove subscription signups.

I don't understand your second yes/no question- is it perhaps clarified by this response?

sean

"upload ads from consecutive magazine issues."
I wonder if there are any copyright restrictions?

Why would I want to view magazine ads online?

I like the idea but not for business practicality reasons. I would peruse / browse the library but I think you'd struggle to get people to pay for the trip down memory lane. Most people would already have a retainer with a designer or advertising agency so that they expect this work is done for them. If you're doin this yourself and design is not your core skill - you need additional funding - IMHO. :)

@siddey - read the idea and comment thread- it's not about "trip down memory lane," it's about extracting actionable insights from past ad campaigns that can be applied to your own advertising. Pure business value here, not leisure/nostalgia.

@Hobbet - i chose magazines because it's a readily-available supply of past visual ads that have run successively and can be quickly acquired into a digital format. i suppose you could photograph billboards over time as well but that's not nearly as easy to do. same thing with scraping online banners that change over time.

@Kevin- I don't know the copyright restrictions here. On first glance it seems the service is providing them with extra exposure of the ads so they shouldn't mind. That show "Firebrand" is an entire show that is just ads (I don't know the deals made behind the scenes with advertisers). Something to research though.

This seems like it could be a wonderful tool across a multitude of industry concentrations. I really like it. When I read your reasoning as to how you arrived at this idea, something I thought of another possible layer of functionality:

If by some non invasive means you could enable tagging on each particular ad(or group of ads), and they were stored categorically(this would take time) and were available for viewing through a searchable database, by name, company, date, publication, artist/model/photographer, and tags. I theorize that besides being able to view, consolidate, and research these advertisements, that you could basically collate and compare the ads that appealed to you and deliver results via a set of filters to expose what commonality if any exists between them, and furthermore, generate additional ads based on your criteria. You might end up with a user driven, automated ad targetting system.

Big Idea, Big Dollars.
Big Dollars

@ideabreaker - yes the tagging feature would definitely strengthen the value of the service allowing people to discover topics of interest and have a richer conversation. it's an excellent suggestion and i would think it would be a feature to be added early in the life of the system. Adding a collaborative filtering element is something else that would be useful- it could be at the level of introducing marketers to other marketers that have similar interest or at the level of the ads themselves.

To sort of clear up what I meant by the last aspect of my post. I just basically see an interface, by which you could insert particular ads up to number (x) which you have stored or you find some sort of affinity for, and basically the system would analyze the ad's by tags, keywords, elements, etc and then begin to offer a series of hit's based on a pretty basic algorithm which would show you a progression(again sortable by category) of similar or progressively evolving ads to pretty much automate the experience which you were sharing about in your initial pitch. Of course it wouldn't be a one-shot, cure-all solution but due to mild degree of randomization a pretty substantial vision could be cast. I feel if not a solution in and of itself, it would be very interesting, fairly engaging, and extremely useful to one sub-set of the populous or another.