Is Your Auto Parts PPC Campaign Working?

A department store owner named John Wanamaker once said “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” John uttered this statement sometime in the 19th century, and despite more than 100 years of innovation and technology in the advertising business, we still don’t have a great system for measuring ads effectiveness…but we’re getting closer.

PPC advertising

If you’re wondering what this has to do with auto parts PPC – and with figuring out how the money you’re spending on AdWords is “backing out” – here’s your answer: You can’t talk about PPC effectiveness until you understand your attribution model.

PPC Ad Performance Is Often Underestimated – Here’s Why

When you buy a click on an AdWords ad (or a Bing ad, a banner, etc.), there are four possible outcomes:

PPC performance image

  1. The customer doesn’t buy. They leave and they never come back. Analytics and AdWords will count this as an unproductive click.
  2. The customer buys the part online right away from your website. AdWords and Google Analytics will count this as an AdWords sale.
  3. The customer loves the part, the price, etc., but they have questions, so they call. After you answer the call and talk to the customer, they order the part from you over the phone. AdWords and Analytics will count this as an unproductive click.
  4. The customer loves the part, the price, etc., but they’re not ready to buy right now. They bookmark your site, leave, and then return to buy the part later. AdWords and Analytics will often count the ad click as unproductive, later attributing the sale to a “direct” traffic visit.

Do you see the problem? Neither AdWords nor Google Analytics will “track” an ad click to a sale in 3 out of 4 scenarios! This is because AdWords and Analytics use “last click” attribution, which means that they’re only counting sales based on the last “thing” the user clicked on before they bought. Here’s why this matters: You can’t figure out if your part PPC advertising is working looking only at the default reports. The default reports use last click attribution, and last click attribution is going to underestimate the impact of AdWords (or whatever pay-per-click ad system you’re using to try and sell parts).

Enter Attribution Modeling

Attribution is about understanding which customer actions lead to a sale. Or, more simply stated, attribution modeling is about giving credit where credit is due. And if you’re advertising auto parts on AdWords, you definitely want to investigate different attribution models before making any big advertising decisions.

The best tool for most AdWords advertisers is the Google Analytics model comparison tool. Assuming that your Google Analytics and AdWords accounts are properly connected, this tool can show you how the impact of your PPC advertising changes using different models (explained below).

  1. First Click Attribution. The first click gets the credit for a sale.
  2. Last Click Attribution. The last click gets the credit for a sale (this is the default attribution model in AdWords and Google Analytics).
  3. Linear Attribution. Each step in the sale process gets an equal share of the credit. If, for example, someone clicks on your AdWords ad, then clicks on a remarketing ad, and then buys, half the credit goes to the initial AdWords ad, and half goes to remarketing.
  4. Time Decay Attribution. This one operates on the premise that purchase intent becomes stronger over time. The first click gets the least amount of credit since it is the oldest interaction. The most recent clicks get more credit, as they are newer.
  5. Last Non-Direct Attribution. This is basically “last click” attribution, except it excludes any sales tracked to “direct” visits. If your site is exclusively relying upon advertising (meaning you have no customer base, brand name, etc.), this could be a good model for you.
  6. Last Adwords Click Attribution. This is basically first click for AdWords and last click for everything else. If a customer has clicked on an AdWords ad, this model will give the credit for any subsequent sales to AdWords. Otherwise, it’s just last click attribution.
  7. Position-Based Attribution Model. This model gives 40% of the credit to the first click, 40% to the last click, and the other 20% to whatever else may have happened between first and last click. This is an increasingly popular attribution model, as it does a decent job of “splitting the difference” between multiple advertising avenues.

Each attribution has it’s pros and cons, so it’s important to look at more than one model whenever you’re making decisions about your advertising. Before you decide to drop AdWords from your parts advertising budget – or invest every dollar in AdWords – you would be wise to check different attribution models.

Evaluating PPC Ad Performance – Tips and Suggestions

There’s no one size fits all attribution model for auto parts PPC (only we do believe first click attribution is probably the safest default choice – more on that below). Here are some points to remember:

1 – Remember tracking technology isn’t perfect. Data is often skewed because users delete cookies, use different devices, block JavaScript, pick up the phone, etc. Your data is always imperfect, and most of the time it’s going to underestimate rather than overestimate the impact of pay-per-click advertising.

2 – Look at more than one attribution model and experiment. Within the model comparison tool in Google Analytics, it’s incredibly easy to see how different attribution models impact the ROI of your PPC ads.

3 – Consider the lifetime value of each customer. Before you make any marketing decision (PPC or otherwise), don’t forget that the lifetime value of each new customer is often much higher than the revenue generated by the first sale you make. If you look at PPC as a tool for growing your customer base in addition to a mechanism for selling parts, you can afford to bid more.

Which Attribution Model is Best For Auto Parts PPC?

Attribution mystery

Before we make a recommendation, let’s look at a hypothetical scenario:

  1. A customer clicks on your PPC ad, visits your site, but does not buy.
  2. One week later, this same customer finds your Facebook page and decides to visit your site again. However, they still do not buy.
  3. A few days later, the customer sees a remarketing ad on a forum. The customer visits your site again, but once again does not buy.
  4. Finally, the customer sees yet another ad, and after visiting your site for the 4th time, picks up the phone and orders a part from you that way.

This hypothetical scenario happens a lot. Between a) the complexity inherent in auto parts (customers often have questions about fitment) and b) the fact that online buyers generally aren’t in dire need (people who need a part ASAP usually source one locally), a lot of parts are sold over the phone. For this reason, we generally suggest first-click attribution. This will inflate the performance of your AdWords advertising as far as your website is concerned, but most of the time that’s appropriate, as phone calls are difficult to track back to an AdWords click.*

*NOTE: We have PPC call tracking systems and utilize them for most of our clients. However, even with call tracking, it’s difficult to know which clicks lead to a call, and which calls lead to a phone sale.

Still, despite the prevalence of phone-based ecommerce sales in the auto parts and accessories industry, the choice of the “best” attribution model is up to you. If you’d like us to help you review your data, our pay-per-click ad management services are very affordable.