Getting Started with Google AdWords: How to Determine Your Test Budget

If you’re considering launching a Google AdWords campaign, then you may be asking yourself, “How much do I need to invest to test this out?  Is $100 enough or do I need thousands?”
It’s an important question and I hear it all the time from prospective clients. Unfortunately, this question is impossible to answer without further research.  For example, we need to answer the following:

1. Which keywords will you be targeting and how much do they cost per click (CPC)?
2. What is the time frame for your test? Do you need to see results in weeks, months, or a year?
3. What is your sales process and typical sales cycle?  Do customers buy the same day they search or does it take months before a purchase is made?
4. What are your typical sales conversion rates?

Let’s go through an example and at the end you’ll know how to estimate a reasonable budget to test a Google AdWords campaign.

Find Your Target Keyword CPCs

In a Google AdWords search campaign, you pay per click.  That means you only pay Google when a prospect clicks on your ad.  If your ad shows up in Google’s results 1,000 times, but no one clicks on it, then you don’t pay a penny.  That’s why AdWords is sometimes called PPC, or pay per click advertising.
So if we’re estimating our budgets, we obviously need to know how much it’s going to cost when prospects click on our ads. And the exact amount you’ll pay depends on the keyword you’re advertising on.  For example, if you advertise on the keyword, “coffee shop,” then you’ll pay a different amount than if you advertise on “mortgage broker.”  Google estimates you’ll pay $2.90 for “coffee shop” and a whopping $13.76 for “mortgage broker.”
That’s a HUGE difference when we’re estimating budgets.
Now, you may be wondering how the heck do you find all the CPCs for keywords.  It’s actually really easy because Google provides you with the Keyword Planner Tool.  Search for your target keywords and the Keyword Planner tool will give you an estimate for how much each keyword will cost per click.  Note that these are estimates so you may pay more or less.

Testing Time Frame

It’s important to realize that you need to go into an ad campaign with a realistic time frame in mind.  As you’ll see later, the time frame will depend on your budget and it also depends on your industry.  There simply may not be enough search volume for your target keywords to get leads and sales data in 1 month.  For example, “mortgage broker” is searched 9,900 times per month in the US.
If 1% of the searchers click on your ad, then you would get 99 clicks from that particular keyword.  Is it realistic that you would get a sale from only 99 website visitors?  Probably not.  Of course, you’ll be targeting more than one keyword.  The goal here is to make sure there is enough search volume for your target keywords to achieve your goals within your time frame.  Plus, if you have a longer time frame, then you can spread out your monthly budget across multiple months.

Your Sales Cycle

This step is easy.  What is your typical sales cycle?  If it’s over 1 month, then obviously you’ll need to test for multiple months to get decent data from a test campaign.  If your customers buy within 1 day, then you know you’ll get almost instant feedback from the campaign once it starts.

Your Sales Conversion Rates

The final step before we can calculate your budget is to use your sales conversion rates.  In my example above, we estimated that we can drive about 99 prospects to our website from the keyword “mortgage broker.”  There are at least two more conversions that need to take place:

1. Prospect has to call, complete a webform, or visit your office after clicking on the ad
2. You have to close the sale

Let’s say your goal is to get the prospect to complete a webform to schedule an appointment.  A reasonable conversion rate for lead generation like this might be 5%.  So out of the 99 visits, about 5 will schedule an appointment.  I’ll assume 100% will show up for the appointment, which is not realistic, but makes this example easier. :)
Next, is your appointment to client close ratio.  Let’s make the math easy here and say you’re sales conversion rate is 20% (and the sales cycle is only a few days) so you would generate 1 new client.

Estimate Your Test Budget

Alright, now it’s time to put all this information together to estimate your test budget.  We already estimated we can generate 1 new client from 100 clicks on a targeted keyword.  Plus, we know our example keyword costs $13.76 per click, so it’s going to cost about $1,376 to generate one sale.
That means we need at least $1,376 for our initial test to get a sale within about a month, based on the CPC, search volume and sales cycle.  So if you only have $100/month to test, then it’s going to take about a year to test just one keyword.  It’s possible you’ll get a sale more quickly, but it’s also possible it will take longer than estimated to get that first sale.  In this example, I would recommend a budget of at least $2,000 to give this one keyword a fair shot.
As you can see, there are a lot of variables that go into estimating your AdWords test budget.  Play around with the Google Keyword Planner to find your target keywords’ CPCs and search volume.  Then run the numbers based on your sales cycle and conversion rates.

Or, contact me and I’ll do it for you! :)

Bing Ads Introduces Performance Comparison Graphs

The Bing Ads UI has long been plagued by obfuscation. Yes, there have been improvements such as the addition of performance trends and change history visualization, but there hasn’t be a way to compare performance data without downloading, formatting  and analyzing reports on your own. Finally, Bing Ads is working on a solution.
Today the company announced it’s working on new performance comparison feature to:

• Compare your current performance to that of a previous period.
• Compare your performance with the MOVING AVERAGE, a reliable performance baseline to measure the success of your campaigns against.
• Compare your performance between weekdays and weekends.
• See performance comparison in the context of change history.

The new feature , now in pilot mode, follows the same layout as the current performance trend with the addition of a “Comparison range” drop down with three options: Previous period, same period last year, moving average.

Both the ability to see performance by Moving average and the shaded Weekend markers are features that are unique to Bing Ads and not seen in Google AdWords.
Moving Average is calculated by averaging the performance metrics across the previous 4 days. All 4 days are of the same day of the week as the original day:

The Moving Average represents the “normal” performance level that can be used as a baseline to measure your daily performance, because it smoothes out the random fluctuations in the comparison trend line, and it also takes care of the “day of the week” factor. It answers advertisers’ frequent question, “how did my campaign do on day x compared with a typical day of the same day of the week?”

The comparative views will be added to the Change History tab as well.
If you’re interested in participating in the pilot, send requests to CHPilot@microsoft.com.

Google Adds Conversion Estimates To AdWords Bid Simulator

Bid Simulator, the AdWords tool that shows click, impression and cost estimates, now includes conversion predictions
.
The Bid Simulator displays estimates for both 1-per-click and many-per-click conversions you might have received with different bids. As with the other estimated data, the conversion estimates are based on clicks from a recent seven day period.



So far, conversion estimates appear to be very limited to high volume keywords. Many accounts may see conversion estimates in the Bid Simulator only for brand keywords. Sparse conversion data clearly makes it hard to make any worthwhile predictions about future performance. Conversion estimates on brand keywords aren’t likely to be useful if the bids are already maxed out (as in the example above).
For those accounts with high converting non-brand keywords, the conversion data could be a helpful signal in determining to whether to test different bids.