Since I've been consulting, I've had the pleasure of doing what I call End To End reviews for several e-commerce clients. In performing an E2E review, I review all online copy and communications from a customer perspective.
As such, I take the role of the customer, typically starting at the home page, moving through the information gathering process, the product catalogue, the purchase process, the fulfillment process, the support process, the follow-up communications and customer marketing.
What I've found without exception is that at some point there is a communications breakdown where the tone or logic in the communications stream breaks down, leaving a disjointed customer experience.
These breakdowns might occur in the sales process (feature-based language on the product page turns into high pressure sales talk on the "buy" page), the fulfillment process (your ecommerce/cart vendor uses dramatically different language or contradicts what your site says up to that point), or the service process (the customer service pages feel like they are from a different company than the rest of the site).
Usually, these disjoints are caused by the fact that these different areas of the site have different "owners" who are ultimately responsible for them. These areas become their own "silos" and exist more or less independently from each other.
While this may makes sense from an internal perspective, it makes zero sense to your visitors.
How do you solve this problem? Well, unfortunately for many companies with long-established business processes, truly "blowing up" the silos is not necessarily an option. The single easiest answer is to appoint a customer advocate within the company who has the sufficient authority to oversee the continuity of c0mmunications. This might be your Director of Marketing or can be someone more junior, as long as they are senior enough (or respected enough) to get different parties within the company to get on the same page.
The tools required to perform an E2E review are not particularly sophisticated. Going through your most common customer processes, "screenshotting" your progress along the way, and making notations as you go is more or less all you need to do.
The first time you do this, I virtually guarantee you'll find and document some disconnects in your communications process that will be sufficient to get stakeholder buy-in if you present the information tactfully.
Alternatively, if you don't have the time to invest or don't have someone who can ignore the trees and see the forest, it's definitely a project worthy of bringing in help for. Either way, I have yet to see a company that does not have some serious breakdowns in their communications process, so it's definitely something you should have a look at.
Tuesday, December 18, 2007
Logic Is Not Always Your Ally
Can you be too smart?
Having worked in several tech companies, I've had the pleasure of working with a lot of very, very smart people (I keep hoping it will rub off but, so far, no dice).
Obviously there's nothing wrong with being smart, but there's definitely a danger in being "too smart."
Here's the deal: when confronted with a marketing challenge, the first temptation is to review the problem, think it over, and come to a highly logical solution. From a logical perspective, you can often come up with a bulletproof position that is fully defensible if questioned.
That said, you may still be dead wrong, as consumers are not necessarily logical creatures (or, at least, they may not share your same viewpoints that lead to your logical answer).
This is critical to keep in mind when preparing a web site, product page or special offer as people will typically NOT do what you expect them to do. Given this fact, it's critical that we take a stance as marketers and make marketing an "outside-in" exercise rather than an "inside-out" exercise.
Think of Jane Goodall. She didn't become an authority on chimpanzees by studying their biology and making sound logical assumptions as to how they would behave. Instead, she went into their environment, observed their behaviors and tested her hypotheses. She took the outside reality and brought it in to her studies, rather than hypothesizing internally and applying it outwards.
Fortunately, as marketers we don't have to go live with primates. We can learn about our markets through research, interviews and surveys. We then need to validate our assumptions through a/b testing and observed behavior so we can do the right thing for our customers, even when it conflicts with our own logic.
Having worked in several tech companies, I've had the pleasure of working with a lot of very, very smart people (I keep hoping it will rub off but, so far, no dice).
Obviously there's nothing wrong with being smart, but there's definitely a danger in being "too smart."
Here's the deal: when confronted with a marketing challenge, the first temptation is to review the problem, think it over, and come to a highly logical solution. From a logical perspective, you can often come up with a bulletproof position that is fully defensible if questioned.
That said, you may still be dead wrong, as consumers are not necessarily logical creatures (or, at least, they may not share your same viewpoints that lead to your logical answer).
This is critical to keep in mind when preparing a web site, product page or special offer as people will typically NOT do what you expect them to do. Given this fact, it's critical that we take a stance as marketers and make marketing an "outside-in" exercise rather than an "inside-out" exercise.
Think of Jane Goodall. She didn't become an authority on chimpanzees by studying their biology and making sound logical assumptions as to how they would behave. Instead, she went into their environment, observed their behaviors and tested her hypotheses. She took the outside reality and brought it in to her studies, rather than hypothesizing internally and applying it outwards.
Fortunately, as marketers we don't have to go live with primates. We can learn about our markets through research, interviews and surveys. We then need to validate our assumptions through a/b testing and observed behavior so we can do the right thing for our customers, even when it conflicts with our own logic.
I Hate Flash Intros and Banners...
OK, so "hate" might be a strong word.
Let me ask you though, are potential customers coming to your web site to be entertained or to gather information? Unless your answer is "to be entertained," don't waste your visitors' time with Flash.
Reasons to avoid Flash:
P.S. If you have a requirement that Flash can help you fulfill, by all means go for it. As a tool it can be brilliant - as a strategy it's horrendous.
P.P.S. Yes, I know there are some great Flash web sites out there. However, they are almost all for brand oriented companies who don't sell online (liquor, soft drinks, etc).
Let me ask you though, are potential customers coming to your web site to be entertained or to gather information? Unless your answer is "to be entertained," don't waste your visitors' time with Flash.
Reasons to avoid Flash:
- While flash intros and pages may win design awards, they lose by being slow and harder to navigate than html.
- Everybody's in a hurry. Why would you make someone wait 20 seconds to get through an animation to find out your message when you could otherwise deliver it instantly? Show your visitors some respect!
- Bandwidth hiccups can make you look stupid. Face it, even on a high speed connection you still get some connectivity ups and downs. Your Flash looks like crap if this happens before it's loaded or during its load time.
- It costs you money. I have yet to see a single a/b test where a Flash home page beat a static page in terms of conversion.
P.S. If you have a requirement that Flash can help you fulfill, by all means go for it. As a tool it can be brilliant - as a strategy it's horrendous.
P.P.S. Yes, I know there are some great Flash web sites out there. However, they are almost all for brand oriented companies who don't sell online (liquor, soft drinks, etc).
Home Page Strategy - Compromise Kills
In building out a home page, many companies try to be inclusive and make compromises to satisfy all stakeholders within their organization.
Avoid this like the plague if you want an effective home page.
I know this sounds harsh, but the truth is that every additional item on a page that does not serve the main goal of the page ends up robbing from that main goal. With this in mind, you definitely want to reserve the bulk of your home page for the one or two main products/services of your organization, relegating the rest to much lesser prominence.
So how do we keep all areas of the organization/site happy and make sure they get appropriate promotion and exposure? Here's some thoughts for different areas:
Customer Support
First off, anyone looking for support will definitely find it as long as you have a clear "Support" link in your top menu. Additionally, give Support a TON of prominence in any purchase confirmations, newsletters and customer communications. After all, nobody needs support until they've started using the product
Smaller Products and/or Sales Channels
"I know sales are still small, but how will we grow sales if we don't put it on the home page?" This is a familiar refrain from product managers the world over.
That said, if a smaller product only contributes 10% of total sales, you'll be losing money like crazy through the opportunity cost of giving it significant prominence on your home page. The solution: a keyword-rich mini-site or landing page devoted 100% to this product. This, combined with significant merchandising of this product to existing customers will garner significantly better all-around results for the product and the company.
Partner and Affiliate Recruitment
A link in your main menu will suffice for qualified partners and affiliates. If they are truly qualified they will have the insight and motivation to click on a menu link without other forms of persuasion being necessary.
Community and Resources
These are areas dear to my heart, but with the exception of specific circumstances (lead generating contests, etc), don't usually deserve major coverage on the home page of an e-commerce site.
Communities and resources, like customer support, should instead be pitched hard to existing customers/members. As well, bear in mind that one of the huge values of community and resource sections is that they are a huge SEO asset, and usually generate the bulk of their new visitor traffic through search.
There are obviously times to break these rules, but don't do so just to take the path of least resistance. If you stay true to these rules and only break them when there's a clear business reason, your home page will be much the better for it!
Avoid this like the plague if you want an effective home page.
I know this sounds harsh, but the truth is that every additional item on a page that does not serve the main goal of the page ends up robbing from that main goal. With this in mind, you definitely want to reserve the bulk of your home page for the one or two main products/services of your organization, relegating the rest to much lesser prominence.
So how do we keep all areas of the organization/site happy and make sure they get appropriate promotion and exposure? Here's some thoughts for different areas:
Customer Support
First off, anyone looking for support will definitely find it as long as you have a clear "Support" link in your top menu. Additionally, give Support a TON of prominence in any purchase confirmations, newsletters and customer communications. After all, nobody needs support until they've started using the product
Smaller Products and/or Sales Channels
"I know sales are still small, but how will we grow sales if we don't put it on the home page?" This is a familiar refrain from product managers the world over.
That said, if a smaller product only contributes 10% of total sales, you'll be losing money like crazy through the opportunity cost of giving it significant prominence on your home page. The solution: a keyword-rich mini-site or landing page devoted 100% to this product. This, combined with significant merchandising of this product to existing customers will garner significantly better all-around results for the product and the company.
Partner and Affiliate Recruitment
A link in your main menu will suffice for qualified partners and affiliates. If they are truly qualified they will have the insight and motivation to click on a menu link without other forms of persuasion being necessary.
Community and Resources
These are areas dear to my heart, but with the exception of specific circumstances (lead generating contests, etc), don't usually deserve major coverage on the home page of an e-commerce site.
Communities and resources, like customer support, should instead be pitched hard to existing customers/members. As well, bear in mind that one of the huge values of community and resource sections is that they are a huge SEO asset, and usually generate the bulk of their new visitor traffic through search.
There are obviously times to break these rules, but don't do so just to take the path of least resistance. If you stay true to these rules and only break them when there's a clear business reason, your home page will be much the better for it!
Friday, December 14, 2007
Email Reputation - Clean Out the Dead!
I read a good article on email reputation the other day by Ralph Wilson at Web Marketing Today. He supplies 11 great tips on improving your delivery rate. In these tips, he rightly advocates removing bounced recipients and finding ways to engage and woo back subscribers, but leaves out one critical tip: clean out the dead!
Particularly if you've been sending email for a long time, it's likely that you have an extremely large percentage of your database that is not interested in receiving your emails. Don't take this personally, it's a natural consequence of time and shifting interests.
"Email is cheap," you might say, "why would I cut people out of my list?" Well, a pattern that I'm starting to see is that if you've been sending emails for a couple years or more, if you're using responsible email practices, and if your open rate is less than 20%, it's likely that 50% or more of your database is effectively dead.
Let me give you a couple examples. I did one project this year where we experienced problems migrating a large list from one email vendor to another. Deliverability was an issue after migrating the database and we needed to act fast. In this case, I filtered the list down so that it included only people who had opened an email in the previous year, had purchased a product in the last 18 months, or had joined the list in the last 3 months.
The result was that we reduced a list of 2.9 million subscribers to a list of only 1.2 million. Sounds scary, doesn't it? When I saw the size of the new list I thought I was in big trouble. That said, a funny thing happened when we deployed the next newsletter.
Although we had cut the list by almost 60%, we still got 90% of the typical response in terms of opens and clicks. While it would have been nice to have kept that last 10%, the huge cost savings of reducing the deployments by 60% and the improvements to deliverability meant that our roi went through the roof and the cost savings more than made it worthwhile (as well as justifying my fees for quite some time).
More recently, I worked on a database about half this size with similar results. The only difference in this case was that without the deliverability crisis we had more time to tinker with the final segmentation rules for who we kept and who we removed from the database.
In this case, we were able to get to a point where we balanced the "age" of the subscriber against their activities in the past 2 years and were able to cut the list size in half while maintaining virtually 100% of the open and click activities as compared to past mailings.
Again, the cost savings and the improvements to deliverability made this a tremendous win and in one fell swoop we virtually doubled the roi on what we were paying to deploy emails.
Bottom line: if you have a large list and have been sending email for over a year then you could achieve some significant cost savings by cleaning up your database.
Particularly if you've been sending email for a long time, it's likely that you have an extremely large percentage of your database that is not interested in receiving your emails. Don't take this personally, it's a natural consequence of time and shifting interests.
"Email is cheap," you might say, "why would I cut people out of my list?" Well, a pattern that I'm starting to see is that if you've been sending emails for a couple years or more, if you're using responsible email practices, and if your open rate is less than 20%, it's likely that 50% or more of your database is effectively dead.
Let me give you a couple examples. I did one project this year where we experienced problems migrating a large list from one email vendor to another. Deliverability was an issue after migrating the database and we needed to act fast. In this case, I filtered the list down so that it included only people who had opened an email in the previous year, had purchased a product in the last 18 months, or had joined the list in the last 3 months.
The result was that we reduced a list of 2.9 million subscribers to a list of only 1.2 million. Sounds scary, doesn't it? When I saw the size of the new list I thought I was in big trouble. That said, a funny thing happened when we deployed the next newsletter.
Although we had cut the list by almost 60%, we still got 90% of the typical response in terms of opens and clicks. While it would have been nice to have kept that last 10%, the huge cost savings of reducing the deployments by 60% and the improvements to deliverability meant that our roi went through the roof and the cost savings more than made it worthwhile (as well as justifying my fees for quite some time).
More recently, I worked on a database about half this size with similar results. The only difference in this case was that without the deliverability crisis we had more time to tinker with the final segmentation rules for who we kept and who we removed from the database.
In this case, we were able to get to a point where we balanced the "age" of the subscriber against their activities in the past 2 years and were able to cut the list size in half while maintaining virtually 100% of the open and click activities as compared to past mailings.
Again, the cost savings and the improvements to deliverability made this a tremendous win and in one fell swoop we virtually doubled the roi on what we were paying to deploy emails.
Bottom line: if you have a large list and have been sending email for over a year then you could achieve some significant cost savings by cleaning up your database.
Monday, December 10, 2007
Strategy - Short term testing vs long term results
Are you as addicted as I am to a/b testing? If so, you need to remind yourself from time to time to step back and look at longer term results.
The reason for this is that an a/b test will always show the "winner" of a short term test, but does not reflect what you're doing for your business in terms of customer loyalty or lifetime customer value (LCV).
For instance, I guarantee you that an aggressive sales email with a discounted offer that expires quickly will always outperform a less aggressive offer. With that in mind, however, an overly aggressive offer will hurt your brand and reduce the trust your best customers have in your offering.
So, how to protect against this? First off, use some common sense! If your home page or emails are starting to look like a multi-level marketing pitch, you've gone too far. Secondly, step back from looking simply at revenues and have a look at other critical indicators, such as the abandonment rate of various pages and the unsubscribe rate from your newsletters.
If your company has the data (which it should if you're serious about staying in business), step back and have a look at average annual sales per customer, as well as month over month, quarter over quarter, or year over year trends.
Remember, just because a single test seems to validate an overly aggressive tactic, that doesn't mean it's the right thing to do.
The reason for this is that an a/b test will always show the "winner" of a short term test, but does not reflect what you're doing for your business in terms of customer loyalty or lifetime customer value (LCV).
For instance, I guarantee you that an aggressive sales email with a discounted offer that expires quickly will always outperform a less aggressive offer. With that in mind, however, an overly aggressive offer will hurt your brand and reduce the trust your best customers have in your offering.
So, how to protect against this? First off, use some common sense! If your home page or emails are starting to look like a multi-level marketing pitch, you've gone too far. Secondly, step back from looking simply at revenues and have a look at other critical indicators, such as the abandonment rate of various pages and the unsubscribe rate from your newsletters.
If your company has the data (which it should if you're serious about staying in business), step back and have a look at average annual sales per customer, as well as month over month, quarter over quarter, or year over year trends.
Remember, just because a single test seems to validate an overly aggressive tactic, that doesn't mean it's the right thing to do.
Wednesday, December 5, 2007
Email Marketing - Don't ask for a password if I'm unsubscribing!
Ever tried to unsubscribe from a newsletter, only to find out you need to enter an account password in order to access the unsubscribe function? Ouch.
Message to marketers - don't ever, ever, ever require this. First off, if the unsubscribe function is being accessed from a newsletter, the technology required to automatically pre-populate your readers' email in your unsubscribe form is so simple that even I could do it!
Secondly, chances are that a majority of the readers who want to unsubscribe probably don't remember what password they originally used for your service or publication. By requiring they log in to unsubscribe, you're pretty much guaranteeing that these folks will flag you as spam instead of trying to figure out their password. So, if you're wondering why your emails are going to the spam folder on Gmail or AOL, this might just be the reason.
Message to marketers - don't ever, ever, ever require this. First off, if the unsubscribe function is being accessed from a newsletter, the technology required to automatically pre-populate your readers' email in your unsubscribe form is so simple that even I could do it!
Secondly, chances are that a majority of the readers who want to unsubscribe probably don't remember what password they originally used for your service or publication. By requiring they log in to unsubscribe, you're pretty much guaranteeing that these folks will flag you as spam instead of trying to figure out their password. So, if you're wondering why your emails are going to the spam folder on Gmail or AOL, this might just be the reason.
Subscribe to:
Posts (Atom)
