Last Click vs First Click and Conversion Attribution

There has historically been a lack focus, knowledge and tools around conversion attribution, but from a personal perspective I always found analysing this information incredibly useful to assess how each channel is performing. This data can then be utilised to help dramatically improve online marketing performance. This view (up until very recently) has generally not been shared in the world of online marketing as the last click has always been (and still is) accepted as the industry standard performance metric. I don’t know why or how this ended up as the default metric, as clearly on bigger purchases (and a lot of small purchases) it’s unlikely a user will convert based on their first visit to a site!

Example user journey

A user does a search or clicks a banner on a website and lands on your website. At this point the user is most likely in the early stages of the buying cycle so just looks at your product/services and prices to see what you have to offer. At this point the user may convert but in reality they will most likely leave your website and look at a number of other sites. A few hours/days/weeks/months down the line the user is now more clued up and had the intent to buy and they will either have a specific site, product or service in mind and as such they will either visit their chosen site directly or carry out another search using more refined keywords. The user then goes onto convert.

Which click actually led to the conversion?

Now there’s always more than one to argue this:

Argument A) The user carried out their initial research based upon the first visit to a site, then later after x amount of searches and/or visits later the user is more certain about the product that they want or they have found the best price. They then refine their search or visit their chosen site directly, the user converts. So for this scenario if your website is not listed when the user is first searching for a product or service then you’ve missed out on chance to gain the sale as your site was never in the running.

Argument B) The conversion from direct referral (last click), now this argument assumes that the users decision making was not influenced by any other online marketing activity and that their decision to purchase/convert was made purely on the basis of the direct referral/last click. When looking at conversion from direct referrals/last click as a whole the stats include all sales that resulted from a direct referral whether this is a direct access, a bookmark referral, PPC advert, organic search listing or a banner ad. The flaw in the last click method of conversion attribution doesn’t take into account of how the user actually found your website in the first instance, which in my mind means that you can’t see how your online marketing activity is performing as you are fundamentally ignoring what channel gained you the sale. The advantage of this attribution method is that its easy to measure, is the default filter on most analytics packages and is still the industry standard method of conversion attribution.

So what’s the difference between first click and last click conversions?

The difference between first and last click conversion will vary depending on the vertical, but taking the following statistics you’ll see that there is a huge difference:

  • Total site conversions for one month (all traffic sources): £53,840
  • Total Organic first click/original referral conversions: £39,510
  • Total Organic last click/direct referral conversions: £9,115
  • Total Paid search first click conversions: £4,228
  • Total Paid search last click conversions: £26,335

From the above stats you can see there is a huge difference between the first click and last click conversions and clearly depending on which one of the metrics you look at will have a huge effect of how you use your online marketing budget. For instance you could be spending far too much on paid search, when that budget could be further invested into SEO and Display Advertising as further investment in these channels will give you the best improvement in your ROI.

Its not just about first and last click

When taking conversion attribution seriously it’s not just about looking at the first and last clicks, you really should be looking at and analysing the activity that takes place between these events so that you can see the complete picture of which of your online marketing channels the user engages with between the first and last events. Once you can see the bigger picture of what led to the conversion, you can then assess how important and what part each of the channels play within the conversion, which ultimately will lead you to improve the way in which you utilise your online marketing budget and improve your ROI.

So where now?

Whilst a great first step in online marketing is to implement conversion tracking so that you can track the ROI from online channels, the basic information provided is somewhat flawed as you are only shown conversions from direct referrals or conversions from original referrals and no data on what happens between first and last click. To get a full conversion attribution picture there are a number of packages that offer it including Kenshoo (although though there are issues in that it doesn’t segregate paid and organic search yet) or Eyeblaster. For a low cost, very basic solution the Google super cookie is an option.

The problem is not just with the lack of interest of the major analytics providers to introduce an all in one solution, but it’s the mindset of the industry as a whole in that most online marketers are happy to go with the “last click wins” mentality, that is until they are shown the real picture.

So why the historic lack of interest in conversion attribution? Taking things to a provider level, the main web analytics packages didn’t and still don’t offer conversion attribution stats and the big media agencies that run PPC and SEO campaign for their clients as loss leaders didn’t show an interest (maybe as it would show online to be more ROI effective than offline). Recently it’s only been the experienced in-house teams and specialist search agencies that have waved the flag for conversion attribution as these are the teams that really understand how important it can be for the success of the overall online marketing campaign. Whilst there is a lot of buzz around conversion attribution it is still in its infancy, so software and tracking will only get better from here on in.

Read more on conversion attribution

For more details on this topic see:

8 thoughts on “Last Click vs First Click and Conversion Attribution

  1. Great post! Really emphasises the importance of taking a step back and looking at the whole picture. As you know, I am a keen advocate of attributing conversions across the channels that were used on the journey… it is the only way to truly optimise!

    We are doing similar stuff at Nedstat with “touch point” reports showing each of the channels/campaigns/keywords used in a users life cycle, aggregated or at an individual level as well as last and first click attribution. I am sure we will see weighted attribution soon too!

    Ben

  2. Good article Dave – I totally agree that the industry is playing catch up here. The irony that digital marketers have prided themselves on accountability versus traditional marketing channels when a lot of them have been counting the wrong numbers!

    There are lots of suppliers out there all with different approaches….expect Google to weigh in here soon ;O)

  3. Dave – you are so correct when you say…

    “Whilst there is a lot of buzz around conversion attribution it is still in its infancy, so software and tracking will only get better from here on in.”

    The future is actually here now — recommend the White Paper @C3Metrics.com

  4. Great article Dave – attribution is the future of analytics. An educated guess is no longer good enough.

  5. This is really a great blog post and shows the real differences that can happen between first and last click wins in plain numbers. I think in the future more and more agencies will also think about solutions that can help them to get a better picture on effectiveness of traffic sources. We’ll see. :)

    This is one of our features in StormIQ, the so called ‘equity share’ model: A sale is not only attributed to the first or last traffic source but to all traffic sources based on your defined weightings. With equity share you are able to assign different weightings to each traffic source and also create keyword groups for both PPC and natural search and assign them different weightings.

  6. Hey Dave,

    Thanks for raising the topic. Attribution gets far less coverage outside of the geek world than it should.

    In addition to first and last click conversion, I’d suggest that linear attribution (each touch point gets credit equally) and weighted attribution (each touch point gets credit, but unequally) also deserve a place in the discussion.

    My point of view has been that attribution is really a religious debate. There is not right or wrong answer, just one that works or doesn’t work for your business.

    For what it’s worth, my company’s product, ClickEquations (http://www.clickequations.com), calculates all 4 types of attribution for your paid search account simultaneously. You can switch between attribution types at any time without losing historical data. You can also compare how different keywords convert under each type of attribution side by side.

    Cheers,
    -Alex Cohen
    Marketing Manager, ClickEquations

  7. Pingback: Why Bidding on Your Brand Name in PPC Advertising is Worth It | ClickAgent Marketing

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>