And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of up

It is often difficult and embarrassing to admit failure and sometimes it is even difficult to see it, even when it is right in front of us. But only by examining our failures can we hope to improve and progress. Hopefully, this article will help others avoid the same mistakes we made.

Keep in mind that web analytics is not always about counting traffic. In fact, that is usually only a small part of it. It is mostly about offering better products and services, improving the website and making each visit to our website a more pleasant experience. It is also about building customer loyalty and confidence.

This incident started when we received a request to cancel web site tracking service for an account. This happens occasionally, but of course, a cancellation is never a welcome sight. Try as we might, we cannot please everyone. So we learn to accept these things; it is just business.

However, it is our policy to investigate every cancellation and try to determine what went wrong. Once someone has decided to cancel, there is nothing that can be done about it. It is too late. Any damage has already been done. We know that we cannot recover a lost account, but we always try to learn something that will help prevent such things happening in future.

The first thing we did that morning was to close the account as requested and issue a credit. We then wrote to the web site owner and informed them that their request had been taken care of. We made no excuses nor did we try to recover the account. But we did ask for help in understanding why they were unsatisfied. We asked a few simple questions as to the reasons for the cancellation and what we could do to improve the service. Our request went unanswered.

Next, we looked up the account details to see what we could learn. We were shocked! This account had been open less than 24 hours! Not even one full day. To be honest, this stung. It was almost personal, a real slap in the face. It was not so much that we had lost an account, but that it happened so quickly. Such a thing had never happened before, so it was a rude awakening.

Once we located the account, we were able to ‘drill down’ to see every aspect of our client’s visits to our web site.

The original visit came from someone searching for a way to monitor traffic on multiple websites. This was indicated by the keywords used in searching the web. In the one day that we did business, they made three visits, looked at 96 pages and spent an average of 14 minutes and 7 seconds on each visit. The average of 26 seconds per page is a bit long, but the 96 pages visited are what really caught our eye.

Bear in mind that that was not 96 different pages, but simply 96 page visits. Some pages were visited several times. Our visitor detail page lists each page in chronological order as it occurs. This lets us see exactly what visitors find of interest and gives us some idea of what is going on in their minds at the time. Keywords and on-page links tell us what subjects were important.

In this case, our visitor went directly to the products page to see what we had to offer. They next looked at the pricing page to see if it was affordable. They returned to the products page using a link that discussed our risk-free, money back guarantee. So we know this was an important consideration.

They then started the purchasing process, but changed their mind and went back to the pricing page for another look. From there, they returned to the buying process via a link that talked about a special offer we had at the time. So now we also knew that the offer had appeal.

In fact, our guest made three attempts at buying the product before finally completing the sale. The very fact that they did buy on the first visit is also a bit unusual. Most people shop around and come back several times before buying. But there was not enough time between visits for this to have been the case. But then again, perhaps they had been shopping around before coming to our site. So this alone was not given much weight. It was just something we noted.

Next, our new client followed the usual procedures: going to the login, changing the default password, setting up the account and looking at the reports. From here, the first signs of confusion and uncertainty begin to show.

Our client next went back to the products and pricing pages. Since no links were used from these pages, we are not sure what they were looking for. But they again returned and logged in and tried to look at statistics. A few minutes later, they returned again to the product and pricing pages for another look.

Finally we began to get a hint of what was wrong. Our client now went to the tutorials and features pages, examining at length an article on analyzing the website data. Then back again to the statistics analysis. So now we begin to see that our client was unsure of exactly how to go about gathering and using the information from their website.

From here their confusion seems to increase. They went again to account setup and then to the help pages. They repeated this process several times over the remainder of the last two visits. Finally, they gave up and cancelled the account.

Our next step was to examine our late client’s site setup parameters. We found that certain fields were not set up correctly, confirming our growing suspicions that our instructions were not clear enough. By looking at the pages visited and examining the subject matter of those pages and links, we know that our visitor became frustrated at not being able to see statistics being gathered in real time from their site.

And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of upd

The first thing we did that morning was to close the account as requested and issue a credit. We then wrote to the web site owner and informed them that their request had been taken care of. We made no excuses nor did we try to recover the account. But we did ask for help in understanding why they were unsatisfied. We asked a few simple questions as to the reasons for the cancellation and what we could do to improve the service. Our request went unanswered.

Next, we looked up the account details to see what we could learn. We were shocked! This account had been open less than 24 hours! Not even one full day. To be honest, this stung. It was almost personal, a real slap in the face. It was not so much that we had lost an account, but that it happened so quickly. Such a thing had never happened before, so it was a rude awakening.

Once we located the account, we were able to ‘drill down’ to see every aspect of our client’s visits to our web site.

The original visit came from someone searching for a way to monitor traffic on multiple websites. This was indicated by the keywords used in searching the web. In the one day that we did business, they made three visits, looked at 96 pages and spent an average of 14 minutes and 7 seconds on each visit. The average of 26 seconds per page is a bit long, but the 96 pages visited are what really caught our eye.

Bear in mind that that was not 96 different pages, but simply 96 page visits. Some pages were visited several times. Our visitor detail page lists each page in chronological order as it occurs. This lets us see exactly what visitors find of interest and gives us some idea of what is going on in their minds at the time. Keywords and on-page links tell us what subjects were important.

In this case, our visitor went directly to the products page to see what we had to offer. They next looked at the pricing page to see if it was affordable. They returned to the products page using a link that discussed our risk-free, money back guarantee. So we know this was an important consideration.

They then started the purchasing process, but changed their mind and went back to the pricing page for another look. From there, they returned to the buying process via a link that talked about a special offer we had at the time. So now we also knew that the offer had appeal.

In fact, our guest made three attempts at buying the product before finally completing the sale. The very fact that they did buy on the first visit is also a bit unusual. Most people shop around and come back several times before buying. But there was not enough time between visits for this to have been the case. But then again, perhaps they had been shopping around before coming to our site. So this alone was not given much weight. It was just something we noted.

Next, our new client followed the usual procedures: going to the login, changing the default password, setting up the account and looking at the reports. From here, the first signs of confusion and uncertainty begin to show.

Our client next went back to the products and pricing pages. Since no links were used from these pages, we are not sure what they were looking for. But they again returned and logged in and tried to look at statistics. A few minutes later, they returned again to the product and pricing pages for another look.

Finally we began to get a hint of what was wrong. Our client now went to the tutorials and features pages, examining at length an article on analyzing the website data. Then back again to the statistics analysis. So now we begin to see that our client was unsure of exactly how to go about gathering and using the information from their website.

From here their confusion seems to increase. They went again to account setup and then to the help pages. They repeated this process several times over the remainder of the last two visits. Finally, they gave up and cancelled the account.

Our next step was to examine our late client’s site setup parameters. We found that certain fields were not set up correctly, confirming our growing suspicions that our instructions were not clear enough. By looking at the pages visited and examining the subject matter of those pages and links, we know that our visitor became frustrated at not being able to see statistics being gathered in real time from their site.

And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of up

In this case, our visitor went directly to the products page to see what we had to offer. They next looked at the pricing page to see if it was affordable. They returned to the products page using a link that discussed our risk-free, money back guarantee. So we know this was an important consideration.

They then started the purchasing process, but changed their mind and went back to the pricing page for another look. From there, they returned to the buying process via a link that talked about a special offer we had at the time. So now we also knew that the offer had appeal.

In fact, our guest made three attempts at buying the product before finally completing the sale. The very fact that they did buy on the first visit is also a bit unusual. Most people shop around and come back several times before buying. But there was not enough time between visits for this to have been the case. But then again, perhaps they had been shopping around before coming to our site. So this alone was not given much weight. It was just something we noted.

Next, our new client followed the usual procedures: going to the login, changing the default password, setting up the account and looking at the reports. From here, the first signs of confusion and uncertainty begin to show.

Our client next went back to the products and pricing pages. Since no links were used from these pages, we are not sure what they were looking for. But they again returned and logged in and tried to look at statistics. A few minutes later, they returned again to the product and pricing pages for another look.

Finally we began to get a hint of what was wrong. Our client now went to the tutorials and features pages, examining at length an article on analyzing the website data. Then back again to the statistics analysis. So now we begin to see that our client was unsure of exactly how to go about gathering and using the information from their website.

From here their confusion seems to increase. They went again to account setup and then to the help pages. They repeated this process several times over the remainder of the last two visits. Finally, they gave up and cancelled the account.

Our next step was to examine our late client’s site setup parameters. We found that certain fields were not set up correctly, confirming our growing suspicions that our instructions were not clear enough. By looking at the pages visited and examining the subject matter of those pages and links, we know that our visitor became frustrated at not being able to see statistics being gathered in real time from their site.

And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of up

Our client next went back to the products and pricing pages. Since no links were used from these pages, we are not sure what they were looking for. But they again returned and logged in and tried to look at statistics. A few minutes later, they returned again to the product and pricing pages for another look.

Finally we began to get a hint of what was wrong. Our client now went to the tutorials and features pages, examining at length an article on analyzing the website data. Then back again to the statistics analysis. So now we begin to see that our client was unsure of exactly how to go about gathering and using the information from their website.

From here their confusion seems to increase. They went again to account setup and then to the help pages. They repeated this process several times over the remainder of the last two visits. Finally, they gave up and cancelled the account.

Our next step was to examine our late client’s site setup parameters. We found that certain fields were not set up correctly, confirming our growing suspicions that our instructions were not clear enough. By looking at the pages visited and examining the subject matter of those pages and links, we know that our visitor became frustrated at not being able to see statistics being gathered in real time from their site.

And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of up

And the most telling feature of all came from examining the client’s web pages. They had never installed the tracking code on the pages so that statistics could be gathered in the first place! And indeed, there were no records from this account in the database.

This made it obvious that although we had sent them the tracking code to be placed on their web pages, we failed to provide guidance that showed exactly how and where to install the code. We failed to explain its importance and how it worked. This was a very basic and stupid mistake.

Most of our clients are pretty tech-savvy, and since we had never had a problem like this before, we had forgotten that not all of our customers are technically proficient. As if that were not bad enough, our whole business was supposed to be focused on the small web sites. It should be obvious that these people were not likely to be technical people.

So what did we learn from all this? First and foremost, our communications were bad. This client never asked for help. But then, no one should have to ask. We should have offered it right from the first, along with an easy link to reach us. We should have made our introduction more personal, giving a sense of friendliness and concern.

We also learned that our tutorials did not answer our client’s questions. Neither did our help files.

Now we have had to re-think our entire presentation and procedures. We have begun a process of updating and rewriting our tutorials and help files, adding video demonstrations and more graphics and examples. We have added help notices in our introductory email to new clients.

Even though it was only one customer out of many, it was a very important customer. They all are, as this quote from Brent Oxley of Hostgator shows:

“One unhappy customer is worth the weight of 1,000 satisfied customers in terms of how they can affect future business, so we strive to keep as many people happy as we can. We may receive 500 letters of praise in a month, but it is that one letter of dissatisfaction that keeps us up at night wondering how we can make things better.”

The process of updating our site is still in progress. It is not an easy job. Some items have been completed, but it will take some time to get around to them all. A website is a constant process of change and improvement. Nothing stays the same for long in the internet world. We made the mistake of getting too complacent. It took a harsh reminder from a dissatisfied client to set us straight.

It is too late to bring that customer back. They are gone forever. But perhaps we can prevent such a thing from happening again. Perhaps you can use our failure to build on your success. I hope so.