What Happens When Your PPC Budget is Cut in Half?

It's the moment any marketer dreads; budget cuts.

This happened a lot to me last year, and I know I'm not the only one. 2009 was a truly horrible year for a lot of folk. But this story does have a silver lining, thanks to the proof in numbers that search marketing can provide.

So getting back to the story; what exactly happened when half of my PPC budget was taken away?

Month 1 – Continue running campaigns as before, at the reduced budget
The first month was the "control" month (aka the "oh shit" month). Nothing was changed in terms of campaigns, adgroups and keywords that were running. The only change was the amount that was spent. This is what happened:

• Clicks through to the web site: Down 53%
• Key actions indicating an intent to purchase: Down 25%
• Online conversions: Down 46%
• Revenue lost (est): 9.5x more than the budget cut

Month 2 – Use the reduced budget to fund top keywords only
To be fair, if your salary was cut in half, you wouldn’t continue to splurge on things. You would take a look at your expenses, downsize, and cut out certain luxuries to try and live within your means.

So in the second month, I focused campaigns back down to basics. The budget was used to fund only the top-performing keywords – those which drove the most traffic and highest quality. As a result, many longtail words went unfunded. This is what happened:

• Clicks through to the web site: Down 61%
• Key actions indicating an intent to purchase: Down 12%
• Online conversions: Down 25%
• Revenue lost (est.): 5.3x more than the budget cut

Month 2 showed some interesting results. When comparing the different approaches it seems that you drive less volume (clicks) when you limit yourself to “head” keywords (this makes sense as you are casting a much smaller net), but as a result you drive better quality results compared to just spreading the budget thinly (as in Month 1).

As a side note, I believe that the Month 2 approach is not a sustainable long-term solution for the reasons highlighted in a previous post: Understanding the Importance of Assist Keywords.

So comparing the two months, Month 2 was better than Month 1. However, that’s not to say that it was good in its own right – especially when the result was still 5x less revenue than the budget that was cut (and we're talking big numbers here).

Month 3 – I got my budget back, plus a bit more
After two months, I was fighting the budget bearers off with sticks. Looking back I like to see this exercise as more of an experiment, even if it was a forced one.

The results proved a point and served as a reminder of how kick-ass search marketing can be when it comes to metrics and accountability. So if you find yourself in this situation, make sure you keep on top of the numbers, and share it with the people who hold the purse strings.

Use Search Marketing to Speak the Voice of the Customer

There are times when internal opinion can often cloud good judgement, especially when it comes to product messaging.

Customer focus groups are a good way to help inform companies when it comes to overall messaging and identifying important customer benefits (not just features that "we" think are cool). But this takes a lot of time and resource, and although rich in qualitative data, it is often limited when it comes to the issue of sample size and statistical significance. On top of this it can be very expensive, even more-so during times when there is barely enough budget to cover basic marketing costs.

Although not a substitute for focus groups, keyword testing is a good way to help inform messaging. It can help validate whether the right words are being used to describe or even categorize a type of product/resource in the language that your customers are using (as opposed to internal jargon or opinion).

A few benefits of doing keyword testing, include:

• High reach, with actionable quantitative results
• Simple to set up and quick to perform
• Easy monitoring (and tweaking if needed)
• Quick, reliable results (two weeks of data is often more than adequate)
• Much less investment compared to full blown customer research (tests I've done have added zero to little additional cost to the overall search program)

Here are a couple of keyword testing scenarios at work:


Scenario 1: "We should change this product category name from x to y because that is what our competitors are calling it"

1. Sponsor both keyword variations in your PPC program. Even better, throw in some additional variations to test also.

2. Monitor the search volume that is being generated on each variation. This will tell you which variation people search on most when thinking about that product category.

3. Check the click-through rate (CTR) and click volume on each variation. The higher the CTR and volume, the more that keyword is resonating with the voice of the customer.


Scenario 2: "Our customer literature descriptions are all over the place - sometimes we call them reports, sometimes guides, sometimes white papers... but they are all the same thing. I vote for just calling everything reports from now on"

1. This scenario screams A/B copy testing. Try taking a PPC campaign focused in the "Learn" phase to test something like this out. For example, one that already calls out a specific call-to-action in the adcopy.

2. Simply swap out the different variations, whilst keeping the rest of the keywords and adcopy the same. For example, "Download your free report" vs. "Download your free guide" vs. "Download your free white paper".

3. See which adcopy version Google serves up the most to learn which version resonates with customers, paying attention to impression and CTR numbers.