Ian Lurie, Portent Interactive, Seattle, WA
How it fits into your Internet marketing strategy
December 29, 2003
Pay-per-click marketing is a great way to get visitors when you need traffic and you need it now. But it's risky: You can spend a fortune, generate many visits, and end up with nothing to show for it. This article will provide you with a high-level view of pay per click marketing, provide some general strategies and provide an example of what to do, and what not to do.
What is Pay-Per-Click Advertising?
On its face, pay-per-click marketing, or PPC, is pretty simple: Search engines and services, such as Google or Overture, provide listings on a per-bid basis. This is in addition to their 'natural' search results, which are still powered by a combination of keywords found on your site, link popularity and other formulae.
If you place the highest bid for a specific keyword or set of keywords, then you rank number one in these paid listings. On Google, , PPC listings show up in the Adwords column on the right-hand side of the screen. Other engines, such as MSNSearch or Yahoo, display P P C listings as 'sponsored listings' in the same column as the natural search results.
If someone clicks on your PPC listing, they arrive at your web site. And you are charged the amount you bid. So, if you bid $.15 per click on 'widgets', and that's the highest bid, you'll show up first in line. If 100 people click on your P P C listing, then the search engine or PPC service will charge you $15.00.
Why PPC Advertising is Bad
But PPC advertising can cost a fortune. It's easy to get caught up in a bidding war over a particular keyphrase and end up spending far more than your potential return. Some PPC engines, such as Overture, offer convenience features such as 'autobid' that will automatically increase your bid amount to maintain a particular rank. That sounds great on its face, but it can get expensive in a big hurry.
Also, ROI can be very hard to measure. Some PPC engines (Adwords and Overture, specifically) provide conversion measurement tools, so that you can track whether your pay-per-click campaigns are generating the desired result. But these tracking tools aren't 100% accurate, and at the time of this writing the smaller PPC providers don't deliver any conversion tracking.
And watch out for junk traffic. Most pay-per-click services distribute a segment of their results to several search engines. While you certainly want your listing displayed on Yahoo, AOLSearch and MSN, you may not want your listings showing up and generating clicks from some of the deeper, darker corners of the Internet. The resulting traffic may look good in statistics reports but is very unlikely to generate a return.
Finally, pay-per-click advertising does not scale. If you get more traffic, you pay more money in direct proportion to that traffic - your cost per click stays constant, and your overall cost increases. Compare that to natural search engine optimization, where you invest a fixed amount of time and/or money to achieve a better rank, and your cost per click goes down as you draw more traffic.
Why PPC is Good
Pay-per-click advertising can generate traffic right away. It's simple: If you spend enough, you can get top placement, and potential customers will see you first. If folks are searching for the keyphrases on which you bid and you've placed a well-written ad, you will get clicks the moment the ad is activated.
So PPC advertising is fast: With some systems, such as Google Adwords, you can generate targeted traffic within a few minutes of opening an account.
PPC advertising is also nimble: Where natural search engine marketing or other forms of advertising can lag weeks or months behind changing audience behavior, you can adjust most pay-per-click campaigns in hours or days. That provides unmatched ability to adjust to market conditions.
PPC can also be a bargain: Sometimes, you can find keyword 'niches' for which the top bid is around $.10 - in that case, PPC is a great option, because you can generate traffic to your site for a fraction of the cost of any other form of paid advertising.
So, balancing the good and the bad, where does PPC fit in? As a focused advertising tool.
The Role of PPC Advertising
Most businesses can't afford to solely rely on PPC advertising. It's too expensive, and bid amounts inevitably climb. But pay-per-click can fill a few important roles:
Campaign- and issue-based traffic: If you have a short-term campaign for a new product, service or special issue, pay-per-click can be a great way to generate buzz. You can start a pay-per-click campaign within, at most, 24-48 hours, and you can generally change the text of your ad in mid-campaign, so adjusting your message is easy. If you need to focus attention for a finite amount of time, PPC is perfect.
Direct-response business: If you sell a product or offer a service that folks can purchase the moment they arrive at your web site, pay-per-click is a great tool. Online stores are a great example: You know that each click generated is a real potential customer, so spending money to increase the number of clicks makes sense.
Niche terms: If you are trying to generate traffic for a highly specific keyphrase, PPC can often provide bargains. For example, you might not want to pay the top bid for 'bicycles', but 'Landshark Bicycles' is probably a lot less expensive ($.10 per click as of this writing, actually).
The overall rule of thumb? Focus, focus, focus. Natural search engine optimization is a PR-based, long-term attempt to grow your brand and image. Pay-per-click advertising, however, should be handled like any other form of paid advertising: Gingerly, and with a clear, quantifiable short- or medium-term goal in mind. In other words, concentrate on conversions, not clicks.
Making it Work: Conversions, Not clicks
How do you engineer a successful pay-per-click advertising campaign? By paying more attention to conversion, and less to clicks. Keep five rules in mind:
If you want to stay on budget, you have to track conversions. What's a 'conversion'? It's any time a visitor to your web site takes a desired action. Examples of conversions might be:
Visitor makes a purchase
Visitor completes a sales inquiry form
Visitor downloads a white paper and registers
A conversion doesn't have to be a sale. But a conversion has to be worth something to you. If you can't think of any measurable, useful outcome of a visit to your site, do not spend money on pay-per-click advertising - there's no point.
Google and Overture provide conversion tracking - most other pay-per-click services do not, so you'll need to look at third party tools such as Urchin, which is also an outstanding site traffic reporting tool, or GoToast, or my own firm's Xed.
If the pay-per-click service you're using doesn't offer a conversion tracker, and you can't afford to pay for a third-party tool, try something more basic: In a spreadsheet, track the number of conversions, total, per day. Do the conversions increase after you start your campaign? If so, you're likely on the right track. If not, then there's very little chance that your pay-per-click investment is working.
Set a Sensible Budget
A lot of folks ask me how much I typically spend on clients' PPC campaigns. My answer is always 'just a bit less than too much'.
A little glib, I know, but the there is no 'right' amount. It all depends on your circumstances. A good formula, though, is:
cost per click is less than:
conversion rate * total clicks * profit per conversion
In other words, the amount you spend per click should always be less than the total profit earned per click. Let's say, for example, that I'm spending $1.00 per click to bring customers to my (totally fictitious) bicycle shop web site. I know that 2% of those visitors contact me regarding products, and that 30% of those potential customers actually purchase something. I also know that I average $10.00 profit on those purchases. Finally, I also know that I get 200 clicks per month. That puts my pay-per-click campaign in this light:
.6% * 200 * $10.00 = $12.00
So, I'm only earning $12.00 per month on my PPC campaign, but it's costing me $200.00. I need to reduce my cost per click, a lot, or cancel the campaign altogether.
Don't make this a hard-and-fast rule, though. While your initial, direct profit from your PPC campaign may disappoint, you might be acquiring loyal customers.
Going back to my bicycle shop example: At this point, I'm ready to cancel my PPC account and never look back. But I dig a bit deeper, and notice that customers acquired from the PPC campaign spend another $800 each, per year, on higher-margin items that deliver an average profit of $200 per sale - I'm getting loyal, long-term business. That changes the picture significantly:
.6% * 200 * $70.00 = $252.00
Suddenly, my PPC campaign is a narrow but definite success. I'm earning $52.00 per month.
If you can't get this kind of precision, pay close attention to your metrics over time: If your sales, leads or other desired visitor actions increased right after you began your pay-per-click campaign, chances are you're on the right track.
But if you're selling a product or service, I strongly recommend that you invest the time and energy to collect this data and crunch the numbers - it will pay off in the long run.
Find Niche Keywords
A lot of folks aim their ads at the broadest possible terms, such as 'dresses', or 'bike parts', or 'search engine optimization'. Since the broader terms get far more searches, it's a strong temptation - with a big disadvantage. Since everyone bids on the broad terms, the cost per click is generally quite high. And the chances of a conversion, even if someone clicks on your ad, is lower.
Focus instead on narrow, focused keywords: 'Bridesmaids dresses', 'road racing tires' or 'Seattle search engine optimization'. These terms will cost less, and searchers who use them will be far more likely to buy.
Google, Overture and most other PPC services will show you estimated cost per click and searches per day for keyphrases - use these tools to test for the best focus, cost and clickthru combination.
Good Writing: Don't Ignore It
Most pay-per-click advertising requires that you write a very short descriptive phrase about your service. Don't underestimate the importance of this phrase - make sure, at a minimum, that your grammar, spelling and overall language is correct and appropriate for your audience. Also verify that your language adheres to the rules enforced by the pay-per-click service - Google, for example, won't allow ads with superlatives ('the best', 'the greatest', etc.), with repeated keywords, or with excessive capitalization.
Play to Come In... Third
Don't pay for a #1 spot, unless you have a good reason.
Most pay-per-click engines require that you place in the top 3 to get placement throughout their distribution network. For example, a top three ranking on Adwords will get you placement on AOLSearch and NYTimes.com, and a top three ranking on Overture will get you placement on MSN Search and Yahoo.
Also, Google Adwords will often bump you up a spot or two if you have exceptional clickthru rates. So, you can bid for a #3 spot, but get a 10% clickthru rate and end up ranked #1. It's a great deal, and there's no point paying for the top spot if you can get it for free.
Finally, a top-3 position will put you 'above the fold' in most users' web browsers. They'll see you the moment they search, and while the number 1 position may have a better chance of getting clicked, my experience is that the top three spots on any given search engine get very similar clickthru rates.
Adjust, Adjust, Adjust: A Corallary
This isn't so much a rule as an overarching concern - don't set up a pay-per-click campaign and then forget about it. You need to monitor your ads on at least a weekly basis. Why?
If search patterns change and your keywords are searched less often, don't immediately alter your campaign - wait at least a few days to make sure you aren't seeing a statistical 'blip'. But keep an eye on things, always, or you might end up spending money unnecessarily. In my experience, a well-designed campaign needs to be 'tweaked' every few weeks.
A Quick Case Study
Planning and running a functional pay-per-click campaign is an art form. Here's an example of one Google ad (modified to protect the innocent) that we edited for a client. Their original adwords spot read:
- Someone might outbid you.
- Or, someone might have dropped out of the top spot, meaning you can reduce your bid and keep a #3 rank.
- Search patterns may have changed.
Low Cost Bicycle Parts
These ads didn't perform well - their ranking, clickthru and conversion rates were very, very poor. Why? Three reasons: First, the ad is far too general - someone searching for a bicycle part on Google will most likely search for the specific part, not for sites that sell everything. Second, the ad doesn't make any strong value proposition - anyone advertising on Google can very likely take my order online, today. Finally, the ad doesn't optimize for the search terms used to find it.
The result? They were paying about $1 per click for a #1 rank, with 800 clicks per day and less than a 1% conversion rate and an average profit per order of $6. No chance of making any profits with that kind of performance:
Order online today
1% clickthru rate
Not good at all. Here's how we changed it. We developed four ads, each focusing on a single keyword combination or group:
1% conversion rate
800 clicks per day
800 clicks * $1.00 per click = $800 cost per day
.01 * 800 * $6 = $48 profit per day
Each ad targets a keyword combination (in the title) that we found is searched more than 50 times per day. A number 3 rank for each ad costs $.15 per click or less. Within a few days, their performance looked like this:
A complete selection, delivered overnight
Shimano STI Component Sets
Overnight delivery on Dura Ace
Tubular Racing Tires
Continental, Michelin, delivered overnight
Phil Wood Bearing Grease
32 ounce jars and cases delivered overnight
12% clickthru rate
The bids we placed earned them a #3 rank, but their high clickthru percentage bumped them up to the #2 or #1 spot for every keyword and phrase (see 'Play to Come In Third', on the previous page, for an explanation).
It was a solid turnaround, built on basic principles: Good niche keywords, solid writing, a smart budget and intelligent placement. By focusing on conversions, instead of clicks, our client got a better result.
Where to Go From Here
Pay-per-click is now a basic Internet marketing tool. Very few businesses can afford to ignore it. But you have to avoid the more-clicks-is-better mentality. Focus on conversion and return on investment, rather than clicks, and you can build a profitable campaign.
PPC Search Engine Listing
In the next few months, I will be publishing a review of my favorite PPC search engines. But for now, you can find a complete listing of PPC engines at Search Engine Watch.
About the Author
Ian Lurie is an Internet marketer in Seattle, WA. He started his web design and marketing firm, Portent Interactive, in 1995. Portent offers complete Internet marketing support, including search engine optimization, e-mail marketing, and web design and development. Recent projects include SEO and production for www.princesslodges.com, SEO, marketing strategy, design and production for www.dessy.com, and, on the more whimsical side, frida.filmateria.com. Ian has a law degree from UCLA and has successfully avoided practicing law for almost ten years.
8% conversion rate
200 clicks per day
Average profit per order: $6.00
200 clicks * $.11 per click = $22 cost per day
.08 * 200 * $6 = $96 profit per day