Site-owners are starting to learn more about the series of sources which people use to visit sites over time.
Will this widespread introduction of data on cross-session attribution lead to some panic reactions, as retailers count the cost of multiple clicks and plead with customers:
“Please DON’T click our ads”
Take a look at this short example of an ecommerce merchant offering a discount if you don’t click on any more paid advertising:
It’s a safe bet that we’re going to see a lot more of this kind of thing as more companies get a glimpse of how many times they may have paid various marketing channels for the same order.
It used to be the case that this information was only readily available if multiple affiliates were claiming commission on the same order (and then only if you had time to actually check), or if you could afford one of the expensive analytics systems.
But in 2011 multi-channel attribution data is becoming much more widely available.
The data is very far from 100% complete. As more of us use more different devices to visits web sites, the chances of all our visits being recorded in one set of cookies is getting more remote. It’s been estimated that each of us has 4 such sets of cookies on the go [source: econsultancy article by Matthew Tod]. Some solutions put a lot of effort into trying to stitch all those sessions together [e.g Intellitracker's single view of the contact].
But there’s no need for that kind of accuracy unless you’re trying to target individuals. What we’re being offered may be sampled, but it’s good enough to confirm what has often been said before: a purchase may come at the end of a series of repeat visits via different channels [source: econsultancy article by Paul Cook]. This is not news. It’s common sense.
Knowing something in theory is one thing, however, and seeing the data from your own site is another. Especially when it’s your budget paying for each of those clicks.
So I can see a rash of attempts like this to try to STOP people clicking on all those links.
But hold on there…
Nothing has actually changed. Our reports are just showing us a bit more about what was already happening.
It has always been like this. Our on-line advertising spend was being distributed in ways which may not have been so easily understood before. But these are the ways which the sophisticated players who could afford the tools and the people to use them already knew about.
- Like the big-budget sites, more of us can now fine-tune our spend in those areas
- We can invest more in trying to get people to the site by means which don’t cost as much as affiliate links and PPC brand advertising
But we’re not going to get people to click and buy in a single visit just because we’d like it to work that way!
I would suggest shifting some of those budgets towards loyalty schemes, email, SEO and social rather than overlays to slap people in the face when they land on the site. The fundamental principles of persuasion and usability still apply.
Off-line we’re all used to running a mix of campaigns through different channels in the expectation (hope!) that people will see several of them, often several times each (!) and finally make a purchase. The same is true on-line. And now more of us have even better opportunities to get detailed insight on this, IF we have the resources to analyse it thoroughly.
So we can manage it better. But we still probably find that we end up paying for several clicks before the final sale.
For my money, the quick return on investment comes from another load of data which Google Analytics is making available right now.
The Site Speed reports are offering insight on something which very few companies bothered to pay for up until now. But site speed matters to customers. Site speed makes a difference to whether real people will even try to buy things from your site as well as whether they manage to do so.
If you want a real competitive advantage over your rivals, I would head for the site speed reports and fix the pages which have problems, while your competition agonise forever over attribution models. You’ll be making your site better for your (probably mutual) potential customers while competitors are squabbling with PPC and affiliate account managers.
The site in the video
Far from being involved in an early pilot on any new Google Analytics features, the site in question is part of the shrinking chunk of the web which does not have GA installed.
It only seems fair to give www.worldofwolf.co.uk a good link. The site looks like a good place to check out if you want to buy Wolfe Garten garden tools in the UK.
As I said in the video: it would be extremely interesting to know if running the overlay has had any effect on the bounce rate. Received wisdom says that something like that overlay with its wall of text would be a real slap in the face. But real life experience often proves ‘received wisdom’ wrong. Maybe the mention of “VIP” in the overlay headline will counteract the complexity of the message.
Site Speed Matters
- Here is an authoritative 2010 post on the subject from Jakob Nielsen, confirming that his findings about site speed from 1997 still hold true: www.useit.com/alertbox/response-times.html
- Here’s some data courtesy of VKI in Vancouver as part of an article about how to track site speed in GA before the function was built in: blog.vkistudios.com/index.cfm/2011/2/15/Your-sites-latency-is-costing-you-in-Revenue-and-InsightGoogle-Analytics-Solution-Reports-Code