Half of All Internet Traffic Is Non-Human
A recent study revealed that 48.5% of Internet activity (pageviews; impressions; clicks; etc.) is conducted by bots – both good and bad.
- Good bots are used by organizations (e.g. Google; etc.) to collect and process vast amounts of information across the web (e.g. index websites for better search results).
- Bad bots are used with criminal intent to defraud others (e.g. impersonate real Internet users to mimic revenue-generating actions).
The latter is problematic for digital advertising as a whole and recruitment advertising in particular.
- The Billion Dollar Problem: AdTech Attracts Fraud
- Case study: Click Fraud in Recruitment Advertising
- 5 Practical Takeaways for Recruiters
The Billion Dollar Problem: AdTech Attracts Fraud
A 2015 US benchmark study shows that untrustworthy supply chains, within the digital advertising space, result in an annual loss of $8.2bn.
The total annual loss is comprised of invalid traffic ($4.6bn), infringed content ($2.5bn), and malvertising ($1.1bn). Recruitment advertising is mostly affected by invalid traffic, which ranges between 6-12% depending on the channel (see figure below).
How Fraud Works:
#1: Finding: Higher bid prices result in more fraud
The goal of fraud schemes is to generate the highest return possible.This can be seen across different types of bidding models, both CPC (cost-per-click) and CPM (cost-per-thousand impressions). A recent report from ANA found that programmatic video advertising is for example most affected by illicit tactics due to its comparatively high CPM rates.
#2: The incentive system informs the tactics
The key success metric (i.e. impression; click; application) of a given campaign dictates which fraud tactics are most effective.
Case Study: Click Fraud in Recruitment Advertising
In July 2016 Perengo partnered up with beta customers from different industries – ranging from hospitality to on-demand jobs. The objective was to programmatically distribute and manage their job ad campaigns via job boards and networks to scale their businesses.
Throughout these campaigns Perengo monitored distribution and conversion tracking data to continuously optimize their performance. The results showed that some campaign traffic received 1 fake click in every 10 clicks. (This number is comparable to Google’s internal AdWords click fraud average of 10%)
This result shows fraud exists in recruitment advertising, and if it had remained undetected would have increased the overall recruitment costs.
This instance was a case of CPC-campaign click fraud, which is primarily caused by revenue-motivated fake clicks or clicks outside the target market (i.e. click farms; bot nets; cheap/redirected traffic). However, a similar type of fraud, which recruiters should be aware of are incentivized traffic and application schemes (e.g. apply for a job and get a $5 gift card) for CPA (cost-per-application) campaigns.
Know your numbers
Before going into detail about how to prevent and manage job ad fraud it helps to have a clear picture of your recruitment process. The cheapest job ad placement is a misuse of resources if it does not produce fully-onboarded hires.
To help hiring managers get a better understanding on their recruitment ROI (return-of-investment) the Society of Human Resource Management (SHRM) published a white paper on a standardized metric: CPH (cost-per-hire)
Recruiters should use this metric to evaluate the development of separate channels over time and make informed budget allocation decisions across their recruitment partners.
A high CPH can indicate:
- Fraud: Lost advertising spend due to bot-generated traffic or incentivized actions
- Targeting: Clicks, which don’t lead to measurable conversions (i.e. applications; hires)
- Channels: Utilizing incorrect channels to reach out to job seekers
- Turnover: Applicants, which become fully-onboarded hires but have a high churn rate within 6 months post hire
5 Practical Takeaways for Recruiters
Fraud prevention and fraud detection is manageable, but if it goes unnoticed it incurs real cost.
In our case, attempted click fraud ranged between 7-10% of all clicks yet we fully recovered the associated cost for our customers. No business should spend more than necessary.
- Test campaigns: Conduct trial runs with recruitment partners.
- Campaign analysis: Monitor your job ad campaigns and application page analytics to detect suspicious behavior.
- Publisher Communication: Ask your recruitment partners to resolve detected fraudulent activity.
- Campaign eCPH: Measure each partner’s performance through a set KPI metric which utilizes campaign cost, application volume, conversions, and quality.
- Specialized partners: If you don’t want to reinvent the wheel, but would like to run optimized campaigns, then work with companies, which specialize in programmatic distribution and campaign performance measurement.
Bottom Line: Monitoring and optimizing your recruitment channels will help you scale your business more cost effectively.
Perengo is a programmatic recruitment tool.
Our goal is to help you grow your business.
Programmatic technology manages your recruitment process and reaches relevant candidates at scale. It monitors performance data along the application funnel to improve results for your business.
Please get in touch with us.
Further Reading and Resources:
- IAB Measurement Guidelines (URL)
- 2015 IAB US Benchmarking Study (URL)
- Google – Ad Traffic Quality (URL)
- Bloomberg – Click Fraud (URL)
- The Bot Baseline: Fraud in Digital Advertising (URL)
- 2015 Bot Traffic Report (URL)
- Wikipedia – Click Fraud (URL)
- Moz – Online Advertising Fraud (URL)
- Confessions of a Fake Web Traffic Buyer (URL)
- Simplicity of Setting Up a Bot (URL)
- SHRM Cost-per-Hire Standard (URL)