Step 1 - Seeing if the Idea is Viable, Testing it out with AdWords

I talked about this first step in the section on Laying the Groundwork - Testing it Out, but now we will see how this works for an actual site - my wife's My Favorite Baby Carrier website.

My wife came across the ERGO baby carrier at a fair one spring, and quickly fell in love with it. She was already on the third baby carrier for our daughter and wasn't happy with the big, bulky frame carrier she was now using on our 25 lb. daughter. The ERGO, in contrast, was a soft, fabric carrier that basically allowed our daughter to ride piggy back.

It was comfortable and both my wife and our daughter loved it. My wife started wearing it everywhere, and often had Moms come up and ask about the carrier - they were intrigued by the look and it seemed so different from the other carriers out there.

After a few months, we talked and my wife just had a real passion for the ERGO carrier. And that is something I look for whenever starting a new web business. So I told her maybe she should sell it on the web - and she really thought that sounded interesting.

She had seen me with my site for a few years, and so she thought it might be fun to have a web business that allowed a lot of freedom while she was a stay-at-home mom. But she was somewhat intimidated by the whole thing - she didn't know where to start or what the whole process would entail.

But she was eager to try, so we started with the basics. Will this product sell on the web? Not every great product will sell on the web. There just may not be enough traffic or interest in the product, it might be a product people need to see and get their hands on before buying, etc.

In our case, we had two questions: was there enough traffic on the web for this kind of product, and would people be interested in this specific baby carrier - one that was pretty neat, but also one that was new and that they would never of heard of or seen before.

The traffic part we could try to figure out. But the second part we just couldn't necessarily know till we had the site up and running. So there are limits to what you can determine before setting up your site - you can't take the risk out of the equation for many products. But you can try to limit it, and we knew that if there wasn't sufficient traffic for these products then we wouldn't want to spend the time on the site in the first place.

Figuring out our Keywords

So we needed to figure out what the traffic was for this kind of product. We got a quick feel for the important keywords by using Wordtracker and the Adwords Keyword Tool .

The Wordtacker site has a free trial, and using the free trial we did a search for the basic keywords relating to baby carriers to see the number of searches for these terms. Wordtracker is a pretty neat tool, and gives you the results for how often these keywords were used in a search over the last 335 million searches over the last 60 days that it has tracked. But I have found that the pure numbers that it gives on searches can really understate the level of traffic on certain keywords. So I use Wordtracker primarily as a way of getting together a good list of keywords.

In this case, I put in the word "baby carrier" in Step 1 of the Wordtracker trial, and left the check marks on for "lateral" and "thesaurus". This is a nice feature of Wordtracker. It will automatically look for related keywords when it is doing its basic keyword search for your term.

I then did this search, and in Step 2 it gave me a list of keywords on the left. By clicking on these, you then see actual keywords with their number of searches on the right. I clicked on a bunch of the keywords on the left, and pretty quickly saw from the detailed list on the right what the basic keywords would be related to "baby carrier".

The ones with the most traffic were general terms "baby carrier", "baby carriers", "baby slings", along with names of specific carriers like "baby bjorn" and "maya wrap". There were a few searches for the term "ergo baby carrier", but it looked like most of the relevant keywords would be more general ones like "baby carriers". My site, then, would need to do a real good job of introducing people to the ERGO carrier since most people coming in would have been just searching in general for "baby carriers" and not necessarily be familiar with the ERGO.

A Bit of Insight Into Relative Sales Potential of a Site from Traffic

I use Wordtracker and the Adwords Keyword tool primarily to get a list of the most important keywords. But I can also get a bit of inkling on the sales potential by analyzing their search results number in a comparative way - I compare the level of traffic to the amount of traffic on keywords that I have a good handle on (usually keywords I use for my business or other clients).

But be forewarned, though, as this example will show, this is a very inexact science and can greatly miss the mark of the actual sales you might receive. In this case, my guess was 1 sale every other day and in a short time we ended up instead getting 2 - 3 sales a day (and later maybe 8 - 10 a day).

The first thing you will need to estimate sales based on the traffic for your keywords is to have a benchmark to use in comparison. I know you won't have this kind of information, so I will give you a benchmark to used.

I had a client that sold videos for Tai Chi for Seniors. I think it is a similar kind of product to the ERGO carrier in that the video would only appeal to a certain percentage of people searching under the keyword "tai chi". Similarly, the ERGO will only appeal to a certain percentage of people searching under the term "baby carrier" because it is a new product that people just haven't heard about so aren't looking for directly.

I don't know if I'm right about this, but that is my take on it - and why I think by comparing the amount of searches for "baby carrier" vs. that of "tai chi" might give me an idea of the type of sales I might expect for the ERGO. I know that I get 3 - 4 sales a day of the Tai Chi for Seniors tape using pay per click advertising - and that Wordtracker tells me that there were about 2700 searches for the term "tai chi". I do check other spellings for "tai chi" to make sure I'm not missing anything, but this seems to be the main term - perhaps a few hundred other searches for other spellings. So maybe 3000 searches total for the term "tai chi" and he was getting 3 - 4 sales a day.

So I then looked at the term "baby carrier(s)", and see where the traffic for the term (counting singular and plural forms) was about 450. Boy this doesn't look good - it is about 1/6 or so the traffic for "tai chi" - so I would be lucky to sell just one every other day. And this wouldn't make the business worthwhile.

A bit disheartened, we then looked at the Adwords Keyword tool to see if it had these same meager results. I've found over time that Wordtracker's results, while helpful in getting a keyword list together, just don't really reflect what you can expect in terms of real world traffic.

I don't know why this is so, but I just know that it can vastly underestimate the amount of traffic you could expect. According to Wordtracker, we could expect 15 people a day to search for "baby carrier", and who knows how many of this small amount would actually click to the site. But before giving up, I wanted to check Adwords Keyword Tool to see if their real world results jibbed with Wordtracker's

And in this case, the Bing Ads results told a somewhat different tale. Adwords Keyword Tool showed almost 6000 searches for the term "baby carrier" in the last month. In contrast, "tai chi" had 46,000 searches. Again the results showed that "baby carrier" was 1/6 or 1/7 or so the amount of searches as "tai chi", but the pure amount of searches per day was much higher - almost 200 times a day Bing Ads had people search for "baby carrier".

This was more encouraging, although I was still concerned by the disparity between the amount of searches of "tai chi" and "baby carrier", and that we still might just expect one sale every other day.

I'm going into detail here on what I thought because I want to show the types of hurdles that you can come to when exploring a site's viability. We almost gave up at this point, thinking we would be lucky to get our one sale every other day.

In fact, as I noted earlier, we started off averaging 2 - 3 sales a day (and got up to 8 - 10 a day later). How could we get the analysis so wrong up to this point?

I think it was by underestimating the value of the ERGO to the people searching for "baby carrier". If you looked at the pure traffic numbers, it wouldn't seem based on these numbers that there would be much of a market for baby carriers on the web. But while the numbers might not be huge, the people searching for this term are very interested, and so are much more likely to buy a product than people using much more general search terms like "tai chi" where they may just be looking for information.

All traffic isn't equal, and in this case the traffic associated with "baby carrier" is very targeted people looking to actually buy. So while not huge, the amount of traffic is so targeted that a baby carrier business can work on the web.

Pure Traffic Numbers Alone Don't Tell the Whole Tale

What is the lesson I've learned from experiences like this (and I've had other similar ones)? The pure amount of traffic for a keyword doesn't necessarily tell you what will work on the web. Some keywords get lots of traffic, but the people searching aren't necessarily interested in buying - they are looking for information.

I learned this in my business - there is huge traffic for keywords like "mattress" or "bed", but not much of this traffic actually is interested in my products. But while there is much less traffic for other more specific keywords, the people using these search terms are actual buyers. So even a small relative amount of traffic for these keywords can sustain a business. In my business, for instance, I do well with a relatively small amount of traffic from more specific search like "visco foam".

So small, targeted traffic can still make for successful web businesses, while large traffic on general terms may not. A bit counterintuitive, and one reason I don't believe in using tools like Wordtracker as a way of finding a niche area for a web business.

While the amount of traffic you will get to your site is important (no traffic means no sales), just as important is the quality of the traffic - and that is something you can only see by doing the site and getting a feel for the conversions (the amount of people who buy vs. the amount of total people who visit the site).

The Next Step - Doing a Test on Google's AdWords

So back to our baby carrier site - we were kind of discouraged at this point with the traffic results from Wordtracker and the Adwords Keyword Tool, but there was enough traffic for the keyword on "baby carrier(s)" to still keep us interested. Luckily, these is one more test we can do that can give us a better feel for the real traffic we could expect.

That is to do real advertising campaign on Google's AdWords for the basic search terms we feel will drive traffic to our site. Since this will entail real costs, we first used the Wordtracker and Bing Ads results to get a general feel at no costs. But now we really want to know the real deal - and for online advertising and traffic AdWords is where the rubber meets the road.

I talked about using AdWords to test out an idea in the Laying the Groundwork section, and I do believe that this most effective tool ever created to get a quick read on your potential market for the web.

Setting up an AdWords Campaign Here's what we did - we set up a new account for my wife at Google's AdWords. It isn't difficult to set up a new account - AdWords does a good job of walking you through the process. As for setting up the actual ad, the basics are pretty easy - although the way AdWords works can get pretty dense. If you have an interest, there are a number of very detailed guides out there on how to get the most out of AdWords. You can find a number of these by just searching for "AdWords" on Google.

But for our purposes, you should be able to set up a basic AdWords campaign pretty easily. The AdWords sign up page has some informational links, as well as a link for a sign up tutorial that is helpful. Once you feel comfortable, click on the sign up button to begin. The first sign up page will ask you to select a language and country for your ad. Select English and United States - the whole country. While you might make a few sales other places, this is your core market and the one you should concentrate on in your test.

Creating an Ad

The second step asks to name an Adgroup and create an ad. The way AdWords is organized, you have a general campaign, and then Adgroups within each campaign you create.

The Adgroup is simply a group of keywords that all are connected to the same ads or ads. So AdWords will first ask you create an ad, then tell it what keywords are to be used for the Adgroup

.

If you want you can create more than one Adgroup, and I do usually create multiple Adgroups when I'm running a real advertising campaign in AdWords. In fact, as you get further with Adwords, having very tightly grouped groups of keywords in separate Adgroups is a big factor in making sure you end up getting high quality scores for your keywords. And high quality scores mean you end up paying less, sometimes a whole lot less, than your competitors to advertise on a keyword.

But for this marketing test, usually you can test out your major keywords by just having one Adgroup with a couple of ads and then a list of keywords.

It is important to do the best ad possible so that your results are as good as possible. I don't claim to be a ad copy maven, but it helps to have the main keyword you are interested in your Headline if possible, and I think it also helps to have the ad copy contain benefits rather than features.

In our test, we came up with this ad copy:

Ergonomic Baby Carrier
Moms love the Ergo Baby Carrier.
Light, comfortable, easy to pack

Our headline included our most basic keyword "baby carrier", and the ad copy concentrated on benefits "light", "comfortable".

If we were going to be creating an actual AdWords campaign for a real website, rather than a test, we would probably end up doing at least one more ad with a different approach. And then compare the relative success of each ad to try to hone in on the best advertising approach. But since this was a test, we just went with this one ad.

Also as part of creating the ad, Google will ask you to put in a display URL and then the web address. For the test, you can just use the URL of the web page you will be sending people to. So find a site that carries your product or service, and use their URL. That way, anyone clicking on your ad will go to a relevant site and not complain to Google about your ad.

Create Keyword List, set a Budget, and Maximum Cost per Click

Once you have created the ad, the next step will ask you to list your keywords and also set a budget. For the test, don't worry about making a comprehensive list of all your keywords. Just hit the big keywords - these will drive most of your traffic anyway.

For our test, we used general terms like "baby carrier" and "baby sling", and also specific terms like "maya wrap" and "baby bjorn" that are product names of popular baby carriers. Since we will be using this combination of general and specific keywords, it is really important that our ad be very specific that it is about the ERGO carrier and the ERGO's benefits. That way we hopefully will only attract clicks that represent people who have a real interest in our specific product.

We could possibly write an ad that would generate more clicks, but this more general ad would not have as many people who are interested in our specific product. And to make our test as good as possible, we need to try to make sure only people really interested in our products click on our ad.

We were then are asked to set a daily budget. It is good to set a high daily budget - if you set one at $20 or so then Google may not show the ad every time someone uses your keyword since Google may ration your ad to make sure it doesn't exceed your daily budget. It is important when doing this test to have your ad run all the time - so set it high. I usually set it at $300 - knowing full well that it won't be nearly this much if I am dealing with these kind of specific keywords. But this ensures that the ad shows all the time.

Once you put in your keywords, AdWords will ask you to set a maximum cost per click. AdWords does allow you later to set a specific maximum bid for any specific keyword if you like, but here it sets up a general high bid for all the keywords you have entered. I generally bid .40 cents to .60 cents to begin with.

I want to come in high with my bed for the test because I want my ad to run in the top three. If it does, it will also show up on AOL and this will increase my traffic count. After you set this, you can check and see where you come in by searching on Google for these keywords and seeing where you ad appears. If this isn't a high enough bid, you may want to bid more for the test. Or if you come in at number 1, you can lower the bid a bit - but not so low you drop out of the top 3 - 4.

Keep an Eye on Your Test

But in any event, once the ads start running you will want to monitor them so you don't spend more than you are comfortable. If you are using these kind of keywords (ones with relatively low levels of traffic) you probably won't have any problems with the costs getting out of control. But if you have a high traffic keyword that gets lots of clicks, keep a close eye on the test and cut it off when you feel you've spent enough. So make sure to set up either warning emails in Adwords when you approach certain spend amounts, or check back often during the day after starting your test.

Once you done this, you can finish up enrolling by filling out all the account information and payment information.

Our Test Results for the Baby Carrier Site

Now, back to our test for the baby carrier site. We ran the test for 2 weeks, and had our ads shown 13,000 times with a bit over a 1% click through rate. So I figured we would get 15 - 20 people a day from this kind of advertising. Add in another 5 - 10 a day we might get from Bing Ads advertising, and I figured we would be lucky to get 30 people a day.

In actuality, though we get 50 - 70 people a day off of AdWords. Why the disparity? I think this is because when we set up the test account, I made a couple of mistakes. I had the general bid at .20 cents. In retrospect, I would have had a more accurate test at .40 cents. At a higher bid amount, the ad would have shown more often since it would have been in the top 3 - 4 shown that show on AOL as well.

I also made one other mistake - I included the keyword "baby carrier" but not the plural form "baby carriers". Now AdWords will automatically do this, but when I ran the ad you needed to put in the plural form. So, I learned the hard way that it is really important to play it as straight as possible to get an accurate test using AdWords. So make sure you come up with the best possible keyword list, bid high enough to make the top 3 - 4, write an ad that is compelling, etc. Only that way will you get a real feel for the traffic.

Back when we did the test, I didn't realize my errors - so we were working on the assumption that we might get 30 people a day. Based on my past experience, I thought that we might then get 1 sale a day or 1 sale every other day.

I asked my wife if this would be worth it to her - maybe 20 or so sales a month. She felt it would be worth it, and in any event was excited to see how it would all work out. And as we have seen, the best test don't necessarily replicate what actually will happen when you do the site.

When you have a site that is actually running, and work the site from all angles (the site side (the effectiveness of the home page, whether people are navigating the site well, the effectiveness of the sales copy, etc) and the traffic side (really working AdWords and Bing Ads to get traffic, getting links, hoping for good search engine placement for your keywords, etc)), you find that you may be able to get a lot more out of the site than the tests indicated.

Hopefully, the test will be more indicative of a floor - a kind of worst case results you can expect from the site. But it may also represent the ceiling as well - so use this information to determine whether the time and costs of putting together the site makes sense.

And -- it is a Go!

In our case, my wife was willing to invest the time and effort for maybe 20 sales a month. And she had my help at no cost which did make this decision much easier. That is one big advantage of learning to design your own sites. It keeps down your costs and allows you to give projects a try that might not seem cost effective. I know I never could have gotten started had I not learned to design my own site - I just wouldn't be able to budget the few thousand dollars I would have needed to get my website started. So self help can give you greater options for exploring site ideas that test marginally as ours did.

Now that the site was a go, we moved on the next step 2, getting the logistics of the site set up - the domain name, the web hosting, 800 number, etc.

Next > Step 2 - getting the logistics of the site set up - the domain name, the web hosting, 800 number, etc.

comments powered by Disqus