What is AI Today
It’s not unusual today to hear people speaking about artificial intelligence (AI). It’s in the media, popular culture, advertising and technology. It just sounds really cool.
At AutoLeadStar, we are often asked, how do we leverage AI and what benefits do our customers receive from this.
First, let's dive into what AI and machine learning actually are, and then we'll list some of the ways our CDXP is infused with it.
Today, AI is a term being applied broadly in the technology world to describe solutions that can learn on their own. These solutions are in most cases advanced algorithms, and these algorithms can look at vast amounts of data and find trends in it; these trends often reveal insights, insights that would be extremely hard for a human to find. Simply put, AI is anything capable of mimicking human behavior, but better.
Now we've established that algorithms capable of mimicking human behavior can be called AI. But if we start to narrow down to those algorithms that can “think” and provide an answer or decision, now we’re talking about a subset of AI called “machine learning.”
Machine learning algorithms apply statistical methodologies to identify patterns in past human behavior and make decisions. These algorithms are also pretty good at predicting; in our industry, we can leverage these algorithms to predict things like the vehicle you may be interested in, the loan terms that would best suit you, and the offers you are most likely to see and then click or convert on.
Leveraging AI across the CDXP
The AutoLeadStar CDXP leverages AI and machine learning in many different ways across the platform and even within specific solutions. Listed below are a few of the ways AI is leveraged at the platform level.
Shopper Identity Resolution
In order to identify shoppers or customers who may come to your website on different devices, during different sessions, and who may even provide different contact details (we all have that old free email or that .edu college email we sometimes still use) you need a way to resolve that identity and merge it into one unified shopper.
With AutoLeadStar, our customers benefit from the machine learning we have developed which helps us to identify multiple shopper entities and merge them into one. This is more than just de-dupping. Taking the devices, contact details, browsing history on your dealer site, vehicles of interest, equity status, conversion details and more - and then unifying all this data into one person who you can then market to with highly specific, personalized, and relevant messaging and offers is one of the foundational ways our CDXP operates. It’s also one of the sophisticated ways we use AI across our platform.
One of the benefits of having connected data orchestration underpin your dealership marketing, is that all industry and market conditions along with shopper intent signals can be pulled into our decision making engines and leveraged across all active solutions on our CDXP.
This is especially relevant when considering budget optimization, that is to say, how efficiently our platform can spend your marketing dollars.
Our unique budgeting optimization algorithm monitors and analyzes internet traffic patterns as well as user conversion rates by channel and platform and is then able to automatically adjust spend and shift channels, in real-time, in order to put funds where you are more likely to target shoppers with higher intent. This means you can sit back while our CDXP shifts between Search, Display, and Social advertising, as well as accounts for any business inputs a dealer may have like targeting limitations or budget allocations for example, in order to get you the steady stream of leads you need to sell your vehicles.
Case in point, throughout 2020 and much of 2021, due to the COVID-10 pandemic, we saw dramatic shifts in internet shopping patterns among people who now found themselves at home more, online more, and using multiple devices to shop. Significant trends in shopper behavior surfaced during this time, which marketers observed and acted on in weeks and months that followed.
However, the AutoLeadStar CDXP was able to leverage our advanced algorithms to uncover these trends earlier and act faster, in real time, so our customers didn’t miss a beat. Funds were automatically gifted from Search networks to Social networks in order to quickly respond to, and in some cases anticipate, new buyer intent signals. We went where the traffic was, where the shoppers were, and where the conversions could take place.
Leveraging AI within our solutions
Listed below are some examples of how AI is leveraged within specific solutions inside our CDXP:
Acquire spends the monthly budget in an organized and completely transparent manner optimizing for maximum LEADS WITHIN A GIVEN BUDGET. As a result, budget shifts take place across the platform and on a near-daily basis as the algorithm is constantly finding the best balance between your spend and getting you the most leads possible (see above for more details).
In order to ensure a smooth and efficient utilization of your budget, Acquire will never spend the entire amount in the first week - that would just be silly. Throughout the month, the budget utilization algorithm makes sure that Acquire spends between 95% and up to 105% of your intended monthly budget.
The algorithm comes into play when you consider that the smooth spend allotment per campaign and per day needs to account for automatic shifts between channels as well as manual inputs from the Dealership. Oftentimes, a dealer will request we add an extra one thousand dollars to the monthly budget on the 15th. Or in some cases, a dealer will need to pull back and lower their total monthly budget on the 15th. Our technology takes these dynamic inputs and makes sure that we are consistently reaching your goals in an evenly metered, organized fashion, addressing your advertising needs in any environment, market, or day of the month.
Multivariate testing is a technique for testing and improving performance in which multiple variables are modified (think A/B testing (human), but with more variables (machine)). The goal of multivariate testing is to determine which combination of variations performs the best out of all of the possible combinations.
Multivariate testing is especially relevant to the ad, email, and engagement templates we create inside our CDXP because we leverage dozens and even hundreds of templates across several solutions. We do this using machine learning and at such a scale, that no human could possibly design, run, analyze, and optimize manually on their own for any of this.
Some places we use multivariate testing include Acquire ad templates. Using dealership specific merge tags like the Dealership brand name, the OEM, city and state, and of course the inventory - we can insert these into hundreds of templates in multiple variations in order to test and optimize the best performing ads.
The same is true of the emails which Nurture generates. Inside Nurture, we’re multivariate testing subject lines in addition to the actual content and offers inside the emails. We do this to ensure that we optimize for the best delivery and open rates as well as clicks and subsequent conversions from the email content.
So while AI is a buzzword you will hear more and more as you surf the web, buy online, or work in most fields tangentially related to the internet, there is warranted excitement about the possibilities it provides and the performance it can deliver.
At AutoLeadStar, we’re sure you’ll benefit from our current and expanding use of advanced algorithms to meet today’s marketing challenges and seize tomorrow’s marketing opportunities for your dealership.