Traffic Up - Sales Down : Where to Look First?

[Planet13]

Hi there, Everyone:

Since the start of Panda in late February, we have seen about a 17% increase in traffic. Yet sales are in a downward slide. What would you look at first to figure out why this is happening?

The majority of that traffic has been going to our better performing products, so we were happy to see the traffic increase. But sales and conversions have just been headed down.

Any suggestions are greatly appreciated.

[toplisek]

Since the start of Panda in late February. What does Panda mean?
Maybe: ---http://www.pandasecurity.com/default

When you see traffic going up to your better performing product check also:
1. Ads targeted locations
2. Demography of visitors who came
3. Install on-line satisfaction query
4. Install eye tracking measure
5. Make more visible FAQ
6. Install search engine (4% of search queries generate 99% Sales! in general testings)
7. Check service again because if you can not make efficient ads to purchase, there is maybe issue with service and its perception.

You see there are number of issues but please give us more information...

[Planet13]

What does Panda mean?

Hmm... I must be spending too much time in the SEO forums. I thought that EVERYBODY knew what Panda was.

Panda is the name of a google algorithm update that happened in the US on February 24th, 2011. It is often called the "farmer update" because it was designed to eliminate Content Farms (sites with lots of low quality content), which it did. Sites like Mahalo and Suite101 reported losing 70% of their traffic due to the update.

My site, as mentioned before, has increased somewhere in the 17% range since then.

[Planet13]

1. Ads targeted locations

I admit that I am horrible at adwords. I need help with that. nearly all my traffic is from search engines, and my overall conversion rate is only around .30% (as in one-third of one percent).

2. Demography of visitors who came

I know I can look up the LOCATION of visitors in google analytics, but don't know how to lookup rest of demographics.

3. Install on-line satisfaction query

do you mean after the sale?

4. Install eye tracking measure

Do you have a suggestion for a good service?

7. Check service again because if you can not make efficient ads to purchase, there is maybe issue with service and its perception.

I am sorry that I don't understand when you say "check service again"?

Thanks in advance for any help.

[scooterdude]

IMHO, its likely that your increase in traffic is made up off non converting queries.

People at a significantly different stage in the buying process, i.e tire kickers

what analytics package do you use, an where you using analytics in the previous period you're comparing with?

[Planet13]

@ scooterdude

IMHO, its likely that your increase in traffic is made up off non converting queries.

That is unfortunately true. However, it seems like it has ALWAYS been that way. Strange thing is that traffic for the queries that SHOULD convert has been increasing too.

People at a significantly different stage in the buying process, i.e tire kickers

Yes, about 2% of visitors will add a product to the basket. But only .3% of visitors will actually buy something.

Part of that is because my traffic is overwhelmingly from google organic. I can't seem to create an adwords campaign that generates sales to save my life.

Is there any "norm" for how many visitors will add a product to their shopping cart?

what analytics package do you use

I have been using google analytics for a while now, so I have some comparison data.

Also, I just started using statcounter since last week (just to "confirm" what google analytics tells me), and my site has some crappy free stats program (only gives you like the top 30 results for each metric). But they both confirm the rise in traffic reported by google analytics.

an where you using analytics in the previous period you're comparing with?

Yup, I will need to drill down deeper to see the change in keyword-specific traffic. I will try to do that tonight.

[toplisek]

1. Ads targeted locations
1. Check also landing pages if you target search engines not Adwords
all my traffic is from search engines...

2. Demography of visitors who came
Make Newsletter, ask people to send their data

3. Install on-line satisfaction query
Make on-line query which shows ALL the time at your landing pages
you quoted after the sale...make always this where you think you are not good in sales

4. Install eye tracking measure
Check: [wilsonweb.com...]
google Eye tracking

5. Check service again. If you can not get sales, there is issue not only in website. It can be also service issue.
What do you sell?

6. Is there any "norm" for how many visitors will add a product to their shopping cart?

This is old way. It is not important how many visitors will add a product. Example:
If my page makes 5.000 USD for each product to be added in shopping cart. Is it important universal measure? I like to know his personal details and call him. Than I make double sales as I talk to him and ask him specific questions. This is the main point. Yes, you have logged users and unlogged. Measure is important for your registered users and logged.

Let me know if it works.

[RhinoFish]

when you sales sales are down, i don't know if you mean gross sales volume or organic sales... and if you mean organic, how are you measuring things...

many things can affect reporting and tracking, i'd start by first understanding the specific issue first.

[Planet13]

@ toplisek:

Thanks for the suggestions and for the explanations. I will try to implement them over the next few days and report back here.

@ RhinoFish

when you sales sales are down, i don't know if you mean gross sales volume or organic sales...

While I mean TOTAL sales, because Adwords is such a minimal part of my sales, I am really concerned about sales generated by organic search engine traffic.

(On another note, I very much want to improve my adwords campaign - but I am seeking help for that in the google adwords forum.)

and if you mean organic, how are you measuring things...

I am using my ecommerce platforms built in stats, as well as using the conversion rate metric displayed in google analytics.

[toplisek]

How do you measure conversion rate? Let us inform as there is maybe also issue.

[RhinoFish]

overall traffic may be up because some content pages ranking increased, or seasonality, or some other factor, point being, looking at overall traffic won't help you diagnose the situation. you need to segment the traffic by source and page and see if the traffic you expect to convert, is converting at the rate you expect it to. segmenting traffic, sales, sources are all important to assembling the puzzle, to make meaning of it, so it can be improved and optimized.

one nice thing about ppc is the speed and flexibility of traffic, and therefore, data collection. think of paid search and organic as partners who help each other. segment the data so you understand each separately. once you reach that level, organic traffic can help you uncover new areas to ppc. and ppc successes can lead to targets for your organic efforts.

[Bewenched]

The first thing I look at is bots and scrapers.

If my sales are down and traffic is high there's usually some scraper copying my content, stealing my sales leads and usually posting a doorway page to Amazon.

[Planet13]

@ bewenched:

...if my sales are down and traffic is high there's usually some scraper copying my content, stealing my sales leads and usually posting a doorway page to Amazon.

Thanks for the warning. Do you have a preferred method for spotting scrapers?

[Planet13]

@ RhinoFish:

Thanks for the input.

...you need to segment the traffic by source and page and see if the traffic you expect to convert, is converting at the rate you expect it to.

That is the strange thing. The CATEGORY / PRODUCT pages that I am targeting ARE the ones seeing the increase in traffic. And the keywords i am targeting are seeing an increase as well.

Again, this traffic is overwhelmingly from google organic (maybe some of it is from google merchant / froogle).

Also, the market I am targeting (the US, where I am located) is seeing the majority of the traffic increase.

[Planet13]

@ RhinoFish:

Again, I really appreciate your help.

..one nice thing about ppc is the speed and flexibility of traffic, and therefore, data collection. think of paid search and organic as partners who help each other. segment the data so you understand each separately. once you reach that level, organic traffic can help you uncover new areas to ppc. and ppc successes can lead to targets for your organic efforts.

sigh... :(

My PPC campaigns suck... They make no money. I probably wouldn't know a good PPC campaign if it came up and bit me on the back bacon.

The ONLY benefit that MY PPC campaigns seem to have is increasing click through rates in my ORGANIC listings in google.

It's as if someone sees my organic listing in the number one or two spot, hesitates to click it, sees an Adwords ad nearby, then decides that I have a "legitimate" site, and clicks through the ORGANIC listing.

If I didn't get this boost, then I would probably stop Adwords, because I don't seem to be doing it very well...

Any PPC resources you can recommend would be much appreciated. I am clueless.

[netmeg]

There's a bunch of things I look at.

First of all, I like to compare apples to apples, and even if the products aren't seasonal, I'm more likely to compare April to last April than April to March.

Ok, so sales are down. But WHAT precisely is down? I had a client come to me complaining that his sales were down since he redesigned his website. We took a look at the analytics, and guess what, his number of sales were way UP, his conversion rate had more than doubled, but his sales DOLLARS were way down. Working backwards, we figured out that the dollar value of his average sale was a little bit more than half of what it was last year. Why? We're still working on that one. Bad economy, obviously. Also he repackaged some of his products - instead of qty 500 to a case, he's now selling qty 100 to a box - so people are buying more boxes. There's more to it than that, but you see where I'm going with this. Figure out exactly where your dropping and by how much, first.

Then you want to look where your traffic is coming from. Take a look at your top keywords for this year and compare them to your top keywords for the same period last year or the year before. Are they the same? Any big changes, plus or minus? Are you suddenly ranking for some really broad search term, or something ambiguous? I get tons of bursty traffic on a small niche affiliate site a couple time a year because the New York Times uses something I rank for as an entry in the crossword puzzle. Talk about useless traffic!

Under the content section, do the same thing with your pages - are the same ones popular this year that were popular last year? Learn to use the advanced segments, so you can get more information about how different types of visitors are using your site.

Do you have ecommerce tracking set up in your Google Analytics? And goals? If so, take a look at the sales funnel. Maybe people are jumping off some place in your shopping cart, indicating you have an obstacle there you didn't realize. Shipping too high, too cumbersome to add the shipping address, could be anything. But you want to look at that.

Personally, I wouldn't worry about doing PPC until you've figured out how to work your Analytics info like a mofo.

There's a ton of Google Analytics resources out on the web - and I'm pretty sure there's gotta be a Dummies book though I've never seen one. But a lot of your answers are in there - you just have to figure out how to fish them out.

[Planet13]

@ netmeg

Thanks so much for the tips and the example. I really appreciate them. I will have to be doing a lot of homework over the next few days to figure out where the individual products numbers rate to last year. I guess this weekend I will be spending lots of quality time with google analytics...

[Planet13]

There is one thing that I am pretty sure about though:

Do you have ecommerce tracking set up in your Google Analytics? And goals? If so, take a look at the sales funnel.

Yes, I do.

The conversion rate once in the funnel is actually only 55.15% (I could have sworn it was higher last time I checked - I thought it was around 80%). While some of the people leaving the funnel go to other pages on my site (and I am sure return back into the funnel, thus skewing the numbers), at each step of the three step process, people do exit the site.

Is a conversion funnel rate of 55.15% a major red flag?

[minnapple]

I work for a few dozen ecommerce sites.
Here are my observations

1. Branded products lines being sold directly by the manufacturer/owner of the brand and these brand owner sites are commanding several top rankings.

2. Big boxes ranking high and putting pressure on pricing points.

3. A continuing decrease in the purchases of high ticket items or multiple items. The dollar amount per purchase is decreasing and is increasing the cost of the acquisition per sale.

[Sgt_Kickaxe]

What would you look at first to figure out why this is happening?

Interesting, here is what I would do.

#1 - Study the data. Break traffic data down by source for starters. "up 17%" doesn't mean much if it's up 50% from one source and down on others. Is your GOOGLE traffic up 17%?

possibilities - Google is sending you more traffic but has pre-filtered what they consider "likely to buy" visitors and sent them elsewhere by showing them personalized results.

#2 - Set up some tests - nothing beats A/B testing to find performance improvements. The trick here is to make small adjustments and allow enough time to pass before making assumptions.

#3 - Which pages is your traffic up on? Perhaps your old performers got a downgrade and your newer content moved up a little.

You'll have your hands full with just those for a while.

[Planet13]

Thanks Sgt_Kickaxe;

You'll have your hands full with just those for a while.

Yes, that is definitely true.

I will sift through the data but off the top of my head:

Is your GOOGLE traffic up 17%?

Yes, organic traffic from google is up 17%.

(My direct / referral traffic is minimal, and my adwords traffic is less than minimal)

possibilities - Google is sending you more traffic but has pre-filtered what they consider "likely to buy" visitors and sent them elsewhere by showing them personalized results.

That is possible. I do have a lot of information - article pages on my site in addition to my ecommerce pages. Although I am seeing a boost in traffic to BOTH kinds of pages. But also the keywords I would associate with ecommerce (in my mind) are up (although the keywords could be used for the information pages as well).

#3 - Which pages is your traffic up on? Perhaps your old performers got a downgrade and your newer content moved up a little.

Actually, old performers are UP across the board. It looks like the well performing products have been holding their own, but the also rans have taken a deeper plunge...

Will post more insights once I get a chance to look at the data some more.

[blend27]

I run a few myself, plus help out as a dev person on about 3 dozen sites. I usually recommend putting a slightly different template/look-n-feel on new ones that don't sell as expected or old stale ones. It's a bit of spice that gets the product going for some reason.

[RhinoFish]

"Personally, I wouldn't worry about doing PPC until you've figured out how to work your Analytics info like a mofo."

Exactly! Especially if your PPC campaigns are, as you said, not good.

Go to a GA seminar:
analytics.blogspot.com/2011/06/summer-time-and-learnings-easy.html

[Planet13]

"Personally, I wouldn't worry about doing PPC until you've figured out how to work your Analytics info like a mofo."

Exactly! Especially if your PPC campaigns are, as you said, not good.

Thanks again for the advice.

I should mention again that they are low risk right now. Not much money is being put into them (under $10 a day, really), but I have seen where they help to boost ORGANIC click through rates. I've done a little bit of A/B testing over the years with that and am fairly confident of the increase in CTR for organic results.

However, I will look at that again over this weekend and see whether it is still worthwhile. It may be where the biggest benefit is reduction in the time spent to manage it will give me more time to work on other things.

[Planet13]

Argghh... this is getting kind of frustrating.

Just went over the numbers again, and traffic is definitely up to my (formerly) main money pages from the markets I target (the US). Most other metrics are all up for those pages, too.

But sales are down.

It looks like I have gained traffic for the broader phrases while the longtail might be dropping, which probably has an affect on sales.

I usually recommend putting a slightly different template/look-n-feel on new ones that don't sell as expected or old stale ones. It's a bit of spice that gets the product going for some reason.

following your advice, I did just redo the layout a few weeks back to make it focus more on the conversion. (I had a few people look at the site and they said it was a little bit confusing). However, that hasn't seemed to help out all.

My funnel conversion rate is 67% (while the overall ecommerce conversion rate is only .31%

I am trying to figure out how I can do A/B testing when my traffic is overwhelmingly organic.

[ecomconv]

I know it's been a while but this is a really interesting thread. The nature of your traffic obviously changed related to Google Panda's release. I think the easy response here is to look at the top keywords in traffic sources you had before and after panda. That should show you that you had more qualified relevant visitors before.
There might also be a change in the effect it had on your PPC campaigns.
Any ideas?

[Essex_boy]

Have you tried to launch a free call service/product phone line ? By useing number specifically for countries I found the sales went up.

[mattia]

DO you know which keywords are used to come to you site?
if not, then this is where you need to start!
Configure and learn how to use Google Analytics to retrieve all the keywords used to come to your site.

Then, if your cart does not do it use a cart that will reveal to you all the keywords that do convert on your site.

if keywords used to come to your site are same before and after the Google Panda, then your conversion issue is obviously on your site (although I doubt this is the case).

my guess is that visitors are now coming to your site using different keywords.

best,
Marc

[edited by: lorax at 12:27 pm (utc) on Oct 16, 2011]

[steveqid]

Google Analytics is not that great rating the conversion for a basic reason. If someone's come to you site and place item in the cart but doesn't check out there is no conversion. Then the same user might come from bookmark and place an order with previously used cart. This way Analytics misses the conversion. I saw it on my site. It took a bit of work to analyze raw logs but I was surprised with scale of my discovery. We're making custom photo id cards so it might be different than other retail but I think the behavior might be common. Analytics should accommodate for this as it's trackable

[edited by: lorax at 12:06 am (utc) on Dec 9, 2011]
[edit reason] no self-links please [/edit]

[Planet13]

@ ecomconv

I think the easy response here is to look at the top keywords in traffic sources you had before and after panda. That should show you that you had more qualified relevant visitors before.

That is a good question and something I will have to look into.

One thing is although we are an ecommerce site with some informational articles, our traffic seems to peek in OCTOBER, and drop slightly during November and December, just when it SHOULD be climbing.

My thought is that google is characterizing us as more of an educational resource than an ecommerce site? Since October is generally the time when educational resources get a boost in traffic...

@ mattia

DO you know which keywords are used to come to you site?
if not, then this is where you need to start!

I will have to double check, but most of the keywords are the same as before (if memory serves). It just seems like they aren't converting. But I will have to double check.

Of course, the fact that google is hiding the keywords for searchers who are logged into their google accounts is going to make it much more difficult to compare.

if keywords used to come to your site are same before and after the Google Panda, then your conversion issue is obviously on your site (although I doubt this is the case).

Well, that is certainly possible. The conversion rate is very low in general. (I think I mentioned it was around .35% of visitors).

One of the truly annoying things is that I get lots of "tire kickers." People who add an item or two to their baskets, and then leave.

Since the items themselves are not easily comparable to other items - most don't have UPCs or SKUs, and if they did, I wouldn't display them anyway - my guess is that they are not really comparison shopping.

my guess is that visitors are now coming to your site using different keywords.

Entirely possible. Will have to look more into that.