Questions of the Week
I spent the last week traveling through the upper midwest. South Dakota to Minnesota to Wisconsin. Ending up in Fish Creek, WI.
Meanwhile, my inbox filled up with comments, questions and progress reports from clients and friends. A couple of trends popped out, so let’s use those as launching points for some quick hit comments. We’ll cover Pricing, Adwords, and Email style formats.
Pricing
First up is pricing. One question was about how I came up with the pricing in my proposal, one was a request for pricing and the other was a friend wondering how to respond to a quote for prices. Essentially, the point behind all the questions was one of Value. How can both parties agree on the value that is provided for the work to be done? The answer to my friend provides the best example of where I’m going with this.
I’ll edit my answer below:
Yes, you can quote him $XX an hour.
The nature of what I’m trying to help my friend with is this: understand what your client is trying to accomplish and you’ll both feel better about what the work is worth. Most of the projects I am approached with don’t end up looking like the initial request because the client and I are rarely talking about the same things. Once we get into the desired results and attach a value to achieving those results, we end up with a new set of possible solutions. We can determine a value from the result and that helps with the investment required.
Get good at answering their questions and then moving toward desired results.
Adwords
Speaking of results, I had a few Adwords questions in the inbox too. How to correctly set up an Adwords Campaign and how to start writing ads. Actually, these weren’t questions as much as they were tweaks to the Google Adwords Recommended Setup. My client took the initiative to set up their own campaign, so my job was easy. Tweak it. They send me a note after a weekend of results and here is the “answer” to questions about the results.
The gist of the setup is this: Google wants you to cast a wide net and learn from the results. Invest heavily up front and then get better. I prefer to be very narrow and locked down, especially on suggested phrases. Unless I know those phrases and terms will convert, it’s best to keep traffic to a trickle. Both approaches work, but in this case we don’t have a giant budget so we need to make assumptions (guess) and learn from smaller data sets.
So, I lock down where the ads show up and I limit the variations of phrases that the ads show up for with delimiters. Then I watch and learn. Which leads to part 2 of the Adwords questions: ads. Why only text ads? Why 3 versions? I’ll get into that here.
I believe that simple searches for text are good indicators of what people are looking for. If I’m looking for a “tropical shirt” I will use some variation of that phrase. To test my hypothesis, I need to isolate that term. Even if it takes a month to get enough data to prove what can be seen as common knowledge. The wide-net approach assumes that there are lots of variations of “tropical shirt” that will give me results I can be happy with. I agree. . .but I need to start with the smallest assumption and move up when my budget is limited.
Now, 3 text ads? Yep. Google will rotate the ads and find the best version, so let’s use this rough outline to start: 60% headline, 30% ad body, 10% visible URL. Since most of the emphasis is on the headline to start, I want to see 3 headline concepts for my set of phrases. For this to work well, those phrases need to be grouped because we’re after relevancy. If I am searching for Tropical Shirts, I expect to see ads that relate to that search phrase.
I try 3 headlines, check back and look to see what’s being clicked on. Rank the results, keep #1, keep #2 and change #3 to a variation of ad #1. Maybe a simple headline tweak or maybe a body tweak. Rinse and repeat. With some low traffic terms, this can take weeks or months to play out but in the end you will be edging your way toward the results we wanted in the first place. The S curve of growth in action. Slow, slow, slow progress and then hockey stick into a plateau.
Email format
Back to starting slow. I love to get some results and then start tweaking. I’m less interested in where we start than what we measure from that start. The debate this week was on my recommendation to use a primarily text based email newsletter format. “People love pictures” says the art director and I don’t disagree. . .but how do you receive your email? Looking at analytics across multiple industries, I can tell you that an increasing # of your customers/clients/prospects are using handheld devices on first view. They see this:
This example is from REI. They do it right. Most phones don’t automatically load the images, so even though this is a beautifully photographed email offer, the wonks at REI assume that I don’t load the pics right away and focus on making it easy for me to see what’s up. Look what happens when I load images:
Without even looking at their analytics I know that click through rates are higher because they took the time to slice and link the image, assuming I wouldn’t see it right away. I will go out on a limb and say conversions went up compared to a control that didn’t have sections/links/text under the images.
So, back to the debate on which email format works best. In your case, I don’t know. I’d start by measuring whatever you’re doing now: opens, click throughs, conversions and then test it against what you’re not doing. (I can go on about open rates for a bit, but I’ll leave that for another day).
Does this help? It seemed to help my friends/clients.
Good stuff.