ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

What Analytics Do You Use?


Last year Jupiter research conducted a study into online advertising. Part of this study included an analysis of what people are measuring. This article will explore their findings and see what lessons we can draw from them.

The survey covered both online advertisers and their agencies, and included companies with turnovers ranging from less than $1 million to those exceeding $500 million. The survey looked at more than 450 companies, and around 160 agencies, mainly (but not exclusively) in the United States. These companies represented just about every industry sector. Questions about web metrics spanned from the use of metrics for campaign planning, to assessment and to performance tuning. In other words, what you do before, during and after an online advertising campaign?

The first thing to note is that it doesn't matter how big you are, or what you do-- the level of expertise in web analytics, and commitment to using them, is pretty much the same. No particular industry stands out as better or worse overall, and being bigger (or smaller) doesn't make you any smarter (or dumber). Advertising agencies are no better or worse at this than their clients.

We've all been told that planning is critical to business success, in any area, not just online. Five percent of companies planning online campaigns do not include any form of analysis at all. The most common was an analysis of what worked in the past (78 percent). As we'll see later, most people are not measuring their campaigns very well when they run them, so it's questionable how useful this is.

The next most common form of "analytics" cited was intuition (69 percent). Barely half (58 percent) bother to look at the stats in their own sites. Audited site metrics, such as those from ABCE, and panel research, such as comScore, rated around 18 percent each.

The critical point here is that 42 percent of people are not looking at their own sites. The primary considerations for planning new campaigns people use are what has been done in the past combined with gut feeling. This is fine if you got it right in the past. However, if your audience is doing something you don't expect, or if behavior patterns are changing, you'll be under-performing. Only by examining what's going on in your site now can you determine what your audience is looking for now, and how appealing your current message is.

What people measure while their campaigns are running is odd, and I'd suggest many people need to push their agencies a little harder. Only 75 percent of people measure how many impressions their ads are getting, while 90 percent measure how many visits these ads generate to the site. It's only by counting both of these metrics that you can determine the clickthrough rate of the ads, which tells you about the quality of work you're getting from your agency. If you break the clickthrough rate down so you're comparing different outlets, you'll learn how their readership compares with the audience you're trying to attract.

Only 72 percent of companies measure how many target actions (purchases, registrations, et cetera) they get from their campaigns. This means 28 percent of companies spending money in online advertising don't know what they're getting for their money.

It is not sufficient to merely count visits to the site. The rate of target actions tells you if your material was exposed to the right audience. A high clickthrough rate but a low conversion rate may indicate your advertising is not reaching the right people. Alternatively, you could be hitting the right people, but there may be problems with your site. The only way to determine this is to measure what happens between clickthrough and target action, to analyze visitor behavior within the site. However, less than half (48 percent) of online advertisers do so. This means that 52 percent have no idea why they are getting the results they see.

Looking at these facts, it's easy to see that the performance of the current campaign is an unreliable basis for planning future campaigns-- insufficient information is available.

In order to survive, a business needs to make a profit. This means it has to spend less than it earns. Less than half (46 percent) measure revenue as a result of advertising, and only half of these measure profit. In other words, most people (73 percent) don't know if their online advertising is costing or earning money. This is one reason why PPC rates in Google are so absurdly high. Obviously we don't have the facts, but my personal experience is most companies are spending much more for their online ads than can be cost-justified. You should know whether you're making a profit or not. If you're making a loss but not doing anything to reduce it, you should be able to explain why that's OK.

Next: Why so few advertisers take steps to improve campaign performance.

Return to Page 1

Given the picture we've outlined, it should be no surprise to discover taking steps to improve campaign performance is rare. Thirteen percent of companies state they do nothing to improve performance of their online advertising at all. These are probably the companies that aren't measuring the performance in the first place.

The good news for sites selling ad space is that less than half (43 percent) of advertisers will drop an under-performing site. It's also good news for agencies that ads that perform badly have a good chance of being left in place, with only 51 percent of advertisers removing them.

Once a visitor lands on your site, you've paid for them, so it's important to retain as many visitors as possible. However, only 55 percent of companies reported they tuned their landing pages. Absolutely no one reported that they made any changes deeper within the site itself as a result of what they learned.

You can often learn what you should do by looking at what others are doing wrong. The picture that emerges from this study reveals some flaws in the way most people undertake online advertising.

The most fundamental gap is the disconnect between getting people into the site and what you deliver once they get there. We all understand that different ads work for different audiences. Most of us now understand that you need different landing pages to retain these different audiences when they arrive. However, once they are inside the site most companies assume a "one-size-fits-all" approach. They may be counting how many people are engaging in the target action, but less than half make any attempt to understand why. Analyzing internal site behavior provides the why.

It's as if there is an attitude that the core of the site is set in stone, never to be changed. Websites should not be regarded as monolithic entities or as static ones. Your site is a set of pages, each one of which has its own pattern of interactions. These interactions vary according to the person who is viewing them and the goal they are seeking to fulfil. The ad and landing page that pulled them into the site tell you what that goal was. Your ad offered them something; it made a promise. The landing page convinced them your site could fulfil that promise. If they didn't make it to the target action, at some point they ceased to believe your site could actually deliver. Analysis of what they did and where they left from tells you when they ceased to believe in your site.

Given that people don't ask why things are happening, it shouldn't surprise us that they're not doing anything about it. Without an understanding of why something is happening, you can't work out what to do to change it. The most important discovery is that so few people ask whether the cost of advertising exceeds the return it produces. Perhaps this is a consequence of not adjusting to the differences between online and offline marketing.

It's impossible to make a direct, simple, correlation between expenditure and profit offline, but it's surprisingly easy online. There's no excuse for not seeking this information. You should know if you're making a profit or loss. You may have good reasons for running at a loss, but you should be able to articulate them.

The patterns we can see here are a call to action for the online marketing industry-- both advertisers and their agencies. We need to improve accountability; we need to accept that it is possible to know what our ad spend is earning; we need to take steps to understand why; and we need to respond to this information.

Brandt Dainow is CEO of Think Metrics, creator of the InSite Web reporting system. Read full bio.

Brandt is an independent web analyst, researcher and academic.  As a web analyst, he specialises in building bespoke (or customised) web analytic reporting systems.  This can range from building a customised report format to creating an...

View full biography


to leave comments.