Thursday, September 29, 2011

Web Analytics in practice: Campaign tracking & offline advertising

[This post is the first post of a new series about Web analytics in practice. The idea is to write very practical posts on various topics based on my own daily experience – as a practitioner. It aims at providing simple tips, advices and examples that – I hope – may inspire and help you – whether you are a beginner or more experienced Web analyst]

We all know how to track online campaigns (banner, SEA, affiliates, social media...) – it has become quite a common practice (if not, you should better get started now!). It is really basic stuff.

However campaign tracking should not be limited to the online world. What about the offline activities that may drive traffic to the online channels? In this post I would like to cover two common offline sources that typically (should) bring traffic to your online properties: friendly  URL’s and Quick Response (QR) codes commonly used in offline ads.

Print ad's & short URL’s
It is very common to use short or dedicated URL’s in print ads (magazine, billboard, brochures...) that are (supposed to be) easy to type such as or (the latest was used in print ads in newspapers and displays). These short URL’s redirect you to specific online content (that often has a much “longer” not-so-friendly URLs).

But how effective are these URL’s?  Are they worth the space they use on the advertising space? Are people really typing these?  Everyone can start arguing – giving his own opinion but the only way to answer the question is of course to MEASURE IT.

It just works the same way as tagging any campaign URL – instead of using the plain URL as destination, configure your redirection using an appended URL with campaign parameters.

For example:
Let’s say I want to use a short URL such as that goes to a new product page Using Google Analytics (but it works just the same way with any other Web analytics tool), I would use the following URL as destination of my short URL:

The advantage of using such common method instead of relying on server stats (that are not always that easy to get) is quite obvious:
  • Availability & centralization: Everything is in one place with all your other online data.Iit gets in your Web analytics tool directly – not need to ask an external provider or IT to get redirect figures.
  • Consistency: it gets measured on same platform and same way as any other traffic sources - you are sure to compare apples with apples.
  • Segmentation: You can leverage the power of segmentation and isolate online behaviour, conversions for visitors coming via this source (and benchmark vs. other segments).
Quite simple and straightforward, isn’t it?

Quick Response codes
In this age of mobile – quick response codes (aka QR codes) are getting everywhere and replacing short URL’s in ads. After all it is much easier to scan a QR code(well, in theory :-)) than typing an URL no matter how short or friendly it is – as long as you have mobile device that can scan such code of course.

In most cases, the QR code is simply added somewhere on the ad or in the brochure for example. Sometimes, some brands are using these in a more creative ways. For example, Victoria Secret in the US leveraged its product “sex-appeal” (or should I the persons wearing the product :-)) and made an intelligent use of QR codes to “encourage” people (well, mostly men actually) to scan them.

Anyway – back to topic - whatever way you are using QR codes in your print material, in the end the important question is: are they driving traffic? Do people really take the time to scan the damned pixel thing or are they just skipping it?

Again the only way to find out is to measure. And it just works the same as short links. The URL that you use to generate the QR code should be a properly tagged URL.

If I want to generate a QR code that will redirect to my blog post on PDCA approach, I will use the following link:

The problem with tagged URL’s is that these can get pretty long and this has a drawback when used to generate QR code: it gets quite dense and may be therefore hard to scan. But there is way to solve this issue: once you have defined your tagged URL, use an URL shortener (such as to get a much shorter URL and then use this one in your QR code generator. You will get a much cleaner QR code – easier to scan  (kudos to Luna Metrics for providing such clever and simple tip!).

Again using such common method for QR codes presents the same advantages as cited previously: centralisation, consistency and segmentation.

Do the right people know about it?
To be honest, tracking such offsite sources is not rocket science and it is quite common sense for whoever has been doing Web analytics. I am certainly not the first one to write about it. But still – it is not something that is systematically done. Why? Because the main hurdle is not at all technical – it is somewhere else.

Most of the time the problem is more a lack of awareness and communication. In many organizations, offline marketing is often managed by different persons than the ones who manage online marketing. Most of the time offline marketing is not aware about the tracking possibilities (and how simple it is) while the Internet marketing is not always involved in the creation of offline content (until the moment they see it – but then it’s too late). So how can it be tracked if the right people don’t know about it?

So it is your role, as a super Web analytics evangelist to inform your marketing organization. Seek out the offline marketing team, explain them what can be done and – most important – the added-value it will bring. But don’t stop there – on the first times, guide them through the process (generate the links or QR codes), show them the results and bring insights until it becomes part of their normal content creation process.

It is not a matter of technology here – but more of process and communications. It’s all about teamwork!

To conclude regarding the use of friendly URL’s or QR code, based on my own experience, I would say that the results were not really impressive (to say the less) – at least as online traffic driver. Maybe you will get better results but then it is up to you to test different approaches, creatives, etc. Especially that as you can measure it – you have no excuse for not testing!

Now it’s your turn now: what is your experience with such offline techniques? What other offline techniques or channels are you also tracking online?

About the post, did you like this first post of the series? Was it useful? Practical or not at all? Please share you feedback. I will definitely take it into account for the next posts.

Related resources & posts:


  1. Very useful article. Thanks for walking us through the specifics of creating URLs that link analytics to offline activities.

    I wonder why you've gotten unimpressive results.

    Perhaps the offline audience is not the type to use a website?

    Or is it situational, the reader isn't in a situation where they can enter a URL or scan a QR code?

    Or the offer is not compelling?

    Look forward to reading the next articles in this series!

  2. Michael

    Thanks so much for this post. Its really interesting to see about the QR code and using a URL-shortener.

    Your point is well taken questioning the success of such efforts. One reason could also be that statistics have become ever more inaccurate as our tests have shown. Nevertheless, do people really take the effort to scan such code .... to visit a website using their smartphone.

    The latest fad is that such billboards now ask people to download an App instead of scanning a code. I wonder how many people are willing to take a picture of an outdoor billboard to download an App. All this to get some information or a discount coupon as this Armani example shows:


    I wonder how many people do this and if 10% do it, will less than 1% continue using the App 30 days after? Does that mean success?

    Michael, thanks so much for sharing. I look forward to your next post.

  3. Nice article, Michael. Even when Twitter is full of GA Premium. :-)

    Everytime I am tracking any offline campaign I am trying to find out how many people will reather type or search for "toyota yaris" instead of typing tagged I am not talking about on-line folks having mobile phone but about those who have PC at home and don't remember the shorter URL. Then it is pretty tough to get proper data since you need to derive traffic from direct / search.

    Esp. when that client don't have stable and significant traffic going to his website.

  4. @David:
    Thanks for the nice feedback. Regarding your question - well it's difficult to answer at this stage. I would need more backing data (maybe in a few months).

    Anyway, current assumptions about low performances could be:

    a)Indeed the target audience - the campaign was very "niche" and maybe these don't especially turn to online to get more info but use traditional offline channels (phone, branches)

    b)The situation may be not appropriate as you mentioned. When they read it - they do not have access to Internet right away (or not inclined to stop what they are doing to type in an URL). By the time, they go to a PC - they don't especially remember the full URL and then type the brand name or brand site - expecting to find info right away

    c)they type the short URL in Google (instead of browser bar) - they get directly to the page (no redirection). I have observed this in search terms reports - but very little volume to be honest

    d)the URL is not friendly enough or not visible enough (creative issue) - they use the normal website URL or Google.

    Was the offer compelling? Apparently yes according to financial results.

    Glad you enjoy the read. Regarding - I don't use its statistics but rely on our Web Analytics stats.

    The problem with QR code is that you need to have a proper QR code reader in your smartphone. You may say that quite simple but you should see how many casual users don't even now how to do it or install a proper one.

    From experience also, sometimes the URL is not correctly redirected or simply doesn't work. You need to retry several times (=> the tip to make these easier to scan).

    In Belgium, mobile data subscriptions are still not that cheap - so maybe people are not always willing to waste part of their data volume in scanning QR code if they don't thing it worth it.

    Even though we tried out QR code on a very specific reader audience (with reasonable share using mobile devices) - results were not really "wow". But again, it may be due to creative, placement... To be tested.

  5. @Vojtech:: I totally agree with your point (see my reply to David) that it is likely that people will simply turn to the brand site (and from there look for what they are after) or Google. In few occasion, they even type the short URL in the search engine.

    As you pointed out - such users are hard to identify - falling under "direct" traffic and being diluted with the rest of the mass.

    Still tracking should be a reflex and common things to do - because even if you don't get huge number, it can still be useful and bring interesting insights about such user segment (assuming you get a reasonable sample size). On top, quality may be there...

    Thanks for your feedback.

  6. Great post Micheal :) !

    I love the idea to track offline marketing campaigns through short URLs and QR Codes.
    Like you said, the hardest is not really to track the URL. The most difficult here is to make the marketing department understand how important it is to add few parameters to the URL. It's really not easy though... Sophie and me tried to do this for few QR Codes last week but it was too late... Hopefully, now they are aware we can track offline advertising and next time they they will ask some help :) !

    Can't wait to read the next !


  7. @Florian Glad you liked the topic of the post. And thanks for sharing your own experience regarding the process / communication aspect. Been there, done that :-)

    As you said, they are now aware but still keep a close eye on this. Do not hesitate to remind them here and there or to check if anything plan on their. Because they may have forgotten about it... Trust me! :-)

    I will do my best to not disappoint you and other readers with the next post of the series.



  8. There's an all-in-one online cloud service, doing everything described by the author. I found it at . It does the job, through just one user friendly dashboard, exactly as I was looking for once I read this article. And it lets you also monitor links you define and their contents (you get an alert if something' gonna change!) I just wanted to inform you guys out there ... probably it solves your issues like it did for me ;-)

  9. Thanks John for the information and for sharing your feedback about Worth checking indeed.


Feedback Form