Slide 1 – Tejas: Thank you for joining us for our Webinar, “Tracking & Attribution: The key to cost efficiency and dynamism in candidate sourcing .” My name is Tejas Deshpande from Adzuna Australia (I work with the business development team., which means if I can’t answer your questions, then I know who can). I am joined by Adzuna Australia CEO Raife Watson. Raife is the person who started Adzuna in Australia and will give you an overview…
Raife … For those of you who don’t know much about Adzuna, we are a smart search engine for job ads that strives to match people to better, more fulfilling jobs faster. We use the latest search technology to scan thousands of websites and bring millions of ads to one place, making the process of finding a job more efficient for candidates. Through our unique model, which promotes speed to market, along with our partnership with The Fairfax Network, Adzuna attracts a higher percentage of passive candidates for recruiters to take to market first.”
Slide 3 – In this webinar we will cover: Tracking & Attribution and why it matters.Methods of tracking and – Models of attribution.
Before I continue, please feel free to ask any questions by sending a message at the side of your screen, Raife and I will attempt to answer your questions at the end of the Webinar.
Slide 4 – Firstly, we know what tracking means, but what about attribution? What is it? Attribution is the method to determine how credit is given to marketing activities for generating conversion activities. This could be in the form of leads, acquisitions or transactions. In the context of staffing – attribution is all about allocating credit to the most important activities responsible generating a job application.
Raife: Why is it so important to track marketing spend Tej??
Slide 5: If you can’t track where your applications came from and don’t attribute them properly then how do you know what works? Or what doesn’t work? When it comes to tracking and attribution, ignorance is not bliss…. It’s damn costly.
If you are spending advertising dollars on a promotion, it should be tracked. If you are spending your staff’s hours on an initiative, it should be tracked. If it drives traffic to your website, it should be tracked to the greatest extent possible.
Slide 6: These companies track where every purchase came from and they are only selling small-ticket items, with low margin: Groceries, t-shirts and electronics. For the staffing industry, finding out how a job candidate, a real person, who could be invaluable for your business, found your business, is an activity worth investing in. Now, let’s take a look at the various methods of tracking.
Slide 7: There are multiple methods for tracking the one conversion. Our goal is to track a job application from discovery through to placement.
Slide 8: We want to know where the candidate started their online journey, in what country and at what time of the day. We can find this out using Google Analytics.
From there, we want to identify the quality and types of applications coming through. This is where a multi-poster like Broadbean, which sits on top of an application tracking system, come in.
Finally, businesses must track how employees treat the applications internally. How long does it take to move that application from discovery through to placement? What is the sales cycle? This is where a CRM like Salesforce can be used.
The overall goal is to reduce the number of steps to find, process and place quality candidates. So, how do we refine, streamline and improve the process?
Slide 9 – Test, test, test!
Do a dummy application from every source via every method they are sending applications to you and make sure they are working properly
If you ever see large spikes or slumps in applications, it’s usually incorrect tracking
If using tracking tags make a parameter generator to ensure that tracking is consistent
Raife: Tej, could you explain the method of how to work out the cost per job application? I imagine this will be crucial in optimising marketing spend and saving money.
Slide 10: Sure can Raife, and yes, working out the cost per application target is vital to your marketing success.
Whilst you may only care about placements, the lead time between advertisement and placement is way too long. Worst case scenario is that the campaign was very poor and you have just wasted months of time and 1,000’s of dollars.
Alternatively, if you are measuring against a cost per application target you would know how it was performing within a week.
Another thing that we tend to find is that when you ask people what their cost per application target their answer is – “as cheap as possible”. The problem with this is that you can only ever get a small selection of candidates ‘as cheap as possible’ and the rest of the candidates are harder to find. This pushes your cost per application up. However if you’ve worked out at what point the application is too expensive, then you know how much is too much at the beginning of your campaign.
Firstly, you need to attach a monetary value to a hire. Of course every hire is different, but you need to determine a benchmark of the value of a hire/the amount you are willing to pay on advertising per hire. This can be changed later if you find that it’s either too high or too low. Next you need to dig into your ATS or analytics and determine historically, on average, how many applications were received per successful hire. It’s important to look at as a large a sample size as you can to ensure that your average is not being skewed by an anomaly. So I’d recommend looking at this over at least a 3 month period, but the longer the time period the better.
So now you have your value per hire and your applications required per hire. All you need to do is divide your value per hire by applications required per hire. For example, say you are prepared to spend $1000 per hire and you usually require 50 applications per hire, then we divide $1000 by 50 and you have a $20 cost per application target.
Slide 11 – Now, onto attribution models. Just to re-cap why it’s important to attribute your spend – If you can’t track where your applications came from and don’t attribute them properly then how do you know what works? Or what doesn’t work?
There are multiple attribution models. So, how do you choose the right model for your business? Ask yourself these questions:
How many marketing channels do you use?
How long is your marketing cycle?
The answers to these questions should help you choose the right attribution model.
Let’s take a look at a scenario and discuss the possible attribution models:
Slide 12 – Day #1: Job seeker Anna wants to find a new job as an engineer. Anna types ‘engineer’ jobs into Google and clicks on one of the organic listings on the Google search engine result page and lands on www.adzuna.com.au.
Slide 13 – Day # 2: Anna continues her search for an engineering job, and clicks on one of the PPC ads on Google to land onto www.adzuna.com.au again. She then subscribes to Adzuna’s job alert to receive daily EDMs about new engineering jobs on Adzuna.
Slide 14 – Day #3: Anna receives an eDM from Adzuna with multiple jobs for engineers. One of the roles looks perfect! Anna clicks on the job and it’s a doozy! She promptly apply’s for the job. Now let’s apply four possible models of attribution to this scenario.
Slide 15– This attribution model focuses on the first marketing channel or touch point that brings the sales conversion. Therefore, the organic search channel gets the credit based on the above scenario example.
Because it gives all the credit on the basis of a single touchpoint, it will naturally overemphasize a single part of the funnel. In this case, the First Touch model overemphasizes the top-of-the-funnel marketing channels that drive awareness.
The First Touch model is also more susceptible than other single-touch attribution models to errors from technological limitations. The issue here is that if you are using conversion tracking (e.g. Google Analytics) in a B2B setting, the time between first touch and the conversion can be longer than the common 30 to 90 day expiration on the tracking cookie. So often times, this model is really attributing credit to the first touch that’s within the cookie expiration window, and not the true first touch.
Slide 16 – This attribution model works on the contrary to the first touch attribution, as the job alert EDM gets the credit for driving in the job application.
By measuring and crediting the entire creation of a sales opportunity based on the last touch, the analytics technology has the smallest time window for an error to occur. In long buyer journeys, which includes the job application process, the time period from last touch to conversion is much shorter, than say, the first touch or the lead touch. This matters because many tracking cookies have a 30 or 90 day expiration. If the conversion doesn’t happen within that window, the marketing channel data will be lost. By attributing 100% of the credit to the last touch, this expiration window becomes redundant because no time lapses between the last touch and the conversion.
Slide 17 – The Time Decay model is a multi-touch model that gives more credit to the touch-points closest to the conversion. It makes the assumption that the closer to the conversion, the more influence it had on the conversion. In this case, the EDM would be allocated the highest amount of credit for the job application, followed by the PPC and then the organic search.
The problem with this assumption is that it will never give a fair amount of credit to top-of-the-funnel marketing efforts because that will always be the farthest away from the conversion.
Slide 18 – The U-Shaped model is a great multi-touch attribution model for marketing teams that focus on lead generation. It’s a multi-touch model that tracks every single touchpoint, but rather than give equal credit to all touchpoints, it emphasizes the importance of two key touchpoints: the anonymous first touch that got the visitor in the door and the lead conversion touch. These two touches get 40% credit each and the remaining touchpoints equally split the remaining 20%.
Using our scenario – 40% would be spent on organic search, 20% on PPC and 40% on EDM marketing.
Whichever attribution model you choose, you are already winning.Simply by thinking about where to allocate your marketing spend in a strategic way, you will make better decisions and improve cost efficiency. Once again, the key to choosing an attribution model for your marketing spend is to ask: How many marketing channels do you use? And – How long is your marketing cycle?
Slide 19 – This concludes our webinar for today. If you have any questions please post them in the text box on the side of the screen.
Thank you for your questions.
Our previous webinars – ‘The Power of Passive Talent’, and ‘How to Create a Seamless Mobile Job Application Experience,’ are both available on Adzuna Australia’s YouTube channel.
My name is Tejas Deshpande….. You’ve been great! Adios Australia.