When Search Marketing Works Quickly Hint-PPC

May 21st, 2010

by Laura Thieme

Often times, I get a new client lead and they tell me they “need help quickly with leads”, which is then followed by the question, “What can you do to help me?”

For some, they think SEO is the answer. But I quickly correct them. That is not the quick way to increase leads. PPC - or paid search campaign management is.

Let me tell you the story of a recent new client who came to me about 6-8 weeks ago.

They’re in the service business. They serve the consumer market. They also have a recent product business, and that business is hurting.

They weren’t getting much previously in terms of leads. They allowed me to review their paid search account in Adwords over the initial phone call. They were a client referral. There was already a level of trust from Day One.

We had the initial meeting. They called me the next morning and asked me to move forward immediately. I “needed to increase leads immediately.”

As of this month, as I looked at their Adwords account, we have brought them 40 accurately tracked leads & online orders, with a lead/online order acquisition cost of approximately $32.

We’re talking next steps about starting a more extensive project for them. So what did we do to increase leads immediately?

1) Pulled data into Bizwatch for historical trend analysis
2) Reviewed and edited campaign settings
3) Changed geo-target settings
4) Changed hours that ads that ran
5) Realized within first couple of weeks that conversions had not been setup accurately; despite their statement they could do it themselves. They made the common mistake of assessing a value with a page view, or a download video. In paid search, there are two real types of conversions, an ecommerce order, or quote request/lead form submission. Not a page view. Define your page views or downloads as Goals in Google Analytics, but not in Adwords.
6) Ensured conversions were tracking accurately.
7) Updated ad copy
8) Reviewed competitors, landing pages, ad copy, pricing
9) Updated ad copy again
10) Reviewed “search query data” in Adwords - LOVE this data
11) Reviewed top converting keywords, with high conversion rates, and lower cost of acquisition (Bizwatch helps with this)
12) Updated keywords
13) Frequent client communication every 2-3 days during initial 3-4 weeks, then weekly after that
14) Measured paid search PPC campaign management success

Results:
1) Went from 1-2 leads per month to 40 quality leads/online orders per month
2) Client has mentioned inventory concerns - ha - love when that becomes the problem - but that is a serious problem we’ve dealt with before - be prepared to handle leads and online orders, and ensure inventory can be addressed rapidly if lead volume increases quickly
3) Cost of acquisition for each lead is very acceptable to the client, but knowing their average online order, I’d like to lower it slightly over the coming months
4) Asking for more business from them - we’re discussing a full contract as a result of our first month of work

Next, we’ll write about what to do when search doesn’t work as quickly as one would hope, or in the form that we’d hope - in terms of lead gen or online sales/ecommerce orders.

Posted in Paid Search Tools, Search Analytics. No Comments »

 

My Client’s Google Adwords Refund for Automatic Matching

November 6th, 2009

by Laura Thieme

I received a response from Google Adwords regarding my request for refunding my client $2,000 for automatic matching to “design school” broad match keyword. My client spent $7600+ in the last three months for this one keyword, out of 20,000 keywords. So the client is spending a good amount on PPC, has two accounts both equal spend, and thus is worth making happy if you ask me. Google indicated they are happy to refund my client $350, out of my $2k request on a $160k/mo ad account. Hmmmmm.

What do you think? I responded and let them know that I’m not happy with $350, as Google decided to show my client for keywords unrelated to “design school” broad match. See below for examples in this post and previous post.

I figured I’d better go back and re-educate myself on keyword match types in Google’s Online Adwords Learning Center. In Google Adwords Learning Center “Keyword Matching Options” it references broad match keyword systems. It now shows you for “homes” in response to search query “houses”. I get that, even though that is NOT how it used to be. Previously, broad match required that I use the one of the words in my search query. So if I really wanted to show up for “houses” in addition to “homes”, I’d have to insert both terms into my broad match string of keyword possibilies. Broad match meant I could show for “new homes”, “homes for sale” and other related queries. So if my client wants “design school” broad match, I’d get keywords that could be in front of “design” or “school” as well as anything in between “design” and “school” and after the word “school”. It meant as long as either design or school was in the query, anything was fair game. That was the “OLD” broad match in Adwords.

Broad match is the default setting as one would read or watch in Google’s Learning Center, which is a huge issue for major campaigns, as well as search newbies. So many people don’t understand match types, nor do they have transparency between agency and business owners/managers to know what type of match types are in place for their PPC campaigns For example, Google states “car repair” will trigger “repair car”. Again, I agree with this. This was the old way of showing ‘broadly matched’ terms. But what about non-related terms as shown below?

Phrase match requires exact sequence “design school” would only allow my ad to show up for something before and after that exact sequence “design school”. Nothing could be inserted in between “design” and “school when using phrase match. But phrase match can be somewhat limiting, as I’ve seen with converting keywords in large PPC campaigns. I like the concept of broad match in addition to phrase match. Sometimes you just can’t guess all the ways someone will search for “design school”. They might put relevant terms in between “design school” and thus I might want to keep broad match.

Exact match requires the same order/sequence of design school, and excludes anything else before or after the sequence for [design school]. So, it turns out hardly anyone searches for just “design school” exact match or Google just doesn’t show us for that.

Negative Match would allow me to get rid of terms I don’t want unrelated to “design school” broad match or phrase match as mentioned in my previous blog post earlier this week. See below. Embedded match is a combination of negative match and exact match. I might have to learn more about that one.

In the Adwords Learning Center it broadly references “homes” might get “houses” searches under broad match, but this can be a really big problem for a client.

In our PPC Conversion Audit for this new client that shows up for “design school”, the following terms have to be closely monitored and determined if relevant under “broad match” settings:

1) “art institute” - Google delivered our ad with the keyword search query “art institute” - relevant or not? - Possibly
2) “make a adenovirus cell for school project” - relevant or not? NOT
3) school floor plans - relevant? NOT
4) football field house designs - relevant? NOT
5) school girl make up for customs - relevant? NOT
6) school skirt sewing pattern - relevant? NOT
7) school hat design - relevant? NOT
8) church design with school & gymnasium 70000 square feet - relevant? NOT
9) school letterhead designs - relevant? NOT
10) cool ties pattern - REALLY? Seriously? cost my client $12 ($11.95)
11) cheerleading uniform design - NOT
12) daycare decoration - NOT - cost my client $11.67
13) nursery decorating ideas - $13.71
14) baby nursery designs - $24.46
15) make an ecosystem - $7
16) zebra pattern school supplies - $12
17) cheap class rings - $11.32
18) brown & green bathrooms - $7.62
19) class of 2013 - $6.72
20) make friends in middle school - do people really search this way? 7.05
21) class of 2012 - $12.41
22) large storybook templates for classrooms - $14.58
23) church blue prints - $14.94
24) perporseful desing mathematics grade one teacher - I’m not sure design school is what they need - maybe first grade spelling would be good here - $13.21
25) school reunion supplies — good root keyword to negative match “supplies” and “reunion”
26) school brochures - $12.92 - good term for negative match - “brochures”
27) designing a sunday school resource room - perhaps a term to negative match - “room” and “sunday school”
28) license display cabinets - this is another one that pisses me off - where is “design school” related to license display cabinets - in any way? “display” related to “design? maybe? this cost $14.44
29) discount class rings - another one that just doesn’t match to “design school” despite class being relevant to school $9.29 for that one

So far, my client has been offered $350 refund out of $2k requested for this one term analysis. I’ll keep you tuned.

The moral of the story - if you broad match your terms, please audit your search terms monthly. Every month, you’ll come up with new negative matches.

Why did I keep “broad match” in Adwords for my client until the other day? Well, Google hasn’t shown their “phrase match” ad for ‘design school’ but four times out of 2,000 impressions. They keep showing the ‘broad match’ keyword phrase. Perhaps now, it will show more often. In addition, look at the terms that converted:

15 keywords out of 500+ keywords converted in the past three months. They were all highly relevant. Some were clear broad match queries but phrase match would help in this particular case. I have phrase match in this keyword case. But oddly enough, Google has only delivered clicks 4 times out of 2,000 impressions, so phrase match is either not showing correctly or not performing at all for us on this term.

Please do not think I’m saying phrase match is the only way you should manage your PPC campaigns. I’m not! Broad match can work, but I’m saying you need to do a PPC Conversion Audit and this is where you will uncover issues like I’ve revealed above. I can save, and you can too, a lot of money for clients by analyzing terms that don’t convert, as well as those that have a high cost of conversion.

Posted in Paid Search Tools, Search Analytics. 3 Comments »

 

Google Adwords Automatic Matching is Now Automatic Like it or Not

November 5th, 2009

by Laura Thieme
Update posted on Nov. 6th re the REFUND

When I first started engaging in PPC campaign management in Google Adwords, I preferred it way over any other PPC ad console. I loved broad match, phrase match and exact match. I learned the value of negative matches. But there is one thing that Google did about a year ago, that I did not like and that was something called “automatic matching”, or what it is now which is what I’ll call “expanded broad matching”. Actually, I’ll call it what it is right up front - it’s a way for Google to make a lot more money without most advertisers being aware of it.

Bizresearch provides PPC Conversion Audits, or a service tied to our Bizwatch Search Analytics tool that looks for non-converting keywords, and high cost keywords that are coming up short in terms of proper ROI/ROAS.

I have a PPC Conversion Audit client that shows up for “design school” not just in phrase match, but broad match as well. I’ve seen value in showing up for large scale campaigns for broad match, or the old way that broad match was used. Broad match meant that either of the two word variance mentioned above (design school) could be used, in any order. Examples: columbus ohio design school, fashion design columbus ohio, columbus design training. Broad match would enable you to monitor how people search, and get ideas to add on under “phrase match”, or in some cases [exact match]. But Google’s expanded broad match, or what at one point was an opt-in system of “automatic matching”, is not such a good idea for its advertising clients. It’s a very good idea for Google to increase revenue without adding relevancy for its users.

Here are the basic details:

Target term: Design School
Conversions - about 16, some of which were for “fashion” related terms
Problem: Cost of conversion exceeds acceptable margins
Additional problems revealed: number of unrelated terms coming under broad match as you’ll see below

Currently broad match and phrase match are in use. Broad gets all the clicks and thus, cost. Phrase match gets basically none of the clicks (4 out of 2,000 impressions)

I do give Google credit on one thing. You can now see most of the search terms that people are entering under their new interface, by selecting a keyword, and choosing “See Search Terms/Selected”. In MOST cases, you can see all the data on keywords with clicks.

Here’s what the new Google Search Query term report shows me, using the new Google Adwords interface. Remember my broad match term is ‘design school’ which means that either the word design or school should be included in the list below, and it can be in any order. Even better if I see the two words together along with other terms that are relevant.

Alternatively, here’s the type of data I’m seeing:

drawing lessons for kids 100.00% $11.44 —- not relevant as this is college level education
vintage skateboard drawing 100.00% $6.24 —- NOT RELEVANT
obstacle courses layouts 100.00% $9.86 — NOT RELEVANT
drawing lessons online 10.00% $13.55 — potentially relevant
mapquest collins school of design 100.00% $8.74 — hmmm, don’t think that’s relevant
nail art designs 100.00% $12.62 —could be relevant
disine your school bus online 100.00% $11.74 — NOT RELEVANT
church design in party supplies 100.00% $11.99 — NOT RELEVANT
class of 010 33.33% $6.92 - EXTREMELY NOT RELEVANT
school donation letter template 100.00% $14.98 — NOT RELEVANT
hair design 100.00% $11.87 -not likely relevant
lesson plan templates 12.50% $9.41 — NOT RELEVANT
church 100.00% $13.26 — huh??? really??? Google??? Seriously, I thought for a second I was using another search engine
make a school map 100.00% $13.96 — STILL SCRATCHING MY HEAD
school bus check designs 50.00% $13.40 — now i’m getting mad
make school work 100.00% $8.77 — still guessing
upgrade cs4 design premium 100.00% $11.22 — sure, somehow not - negative match opportunity on cs4 right, upgrade another possible negative term
educational value in mexico grafic 100.00% $8.58 —- laughing on this one
nursery design 50.00% $12.90 — OKAY, NEGATIVE MATCH IDEA “NURSERY”
classroomarchitect 100.00% $13.55 - hmmmm
2d design class 50.00% $4.71 — don’t think so
class ring design 50.00% $10.06 - negative match idea ‘class ring’
dishware school design 50.00% $12.31 - hmmm, maybe
school project desing map 50.00% $12.02 - no
create your own back to school backpack 50.00% $9.70 - negative match idea

None of the keywords above have ever had a conversion. They’re all tied to the keyword “design school” broad match. Yet, in some of the cases it’s because we just need to add a negative match. For example, to get rid of anyone searching on “class ring design”, I can add “ring” or “class ring” to my negative keyword list. I can add “backpack” and “back pack”.

But why is my client showing up for “class of 010″ and “class of 2009″ - how does that relate to “design school”? The other two columns of data show CTR click thru rate and Cost Per Click. Yes, my client is pay $12.02 for the keyword “school project desing map” and $9.70 for “create your own back to school backpack”. Typos included. Some of these terms can be added, like “back pack” as mentioned above. But class of 2010, doesn’t have either word “design” or “school” in it. It’s related to “school”, and this is part of Google Adword’s automatic matching.

According to my Google team, they say that it’s automatically turned on for broad match. I can’t turn it off anymore, even if I wanted to. I’d have to turn off the design school broad match keyword. I keep adding more negative terms, but Google keeps coming up with new ones. Noting the 50-100% CTRs, and the high cost per click, I’ve opted to pause this term.

In case you’re wondering, why don’t I just put this term as phrase match “design school” and “design schools”? Well, I have had it on for a couple of months, and do you know that Google served my client 4 clicks for that variance. Over 2,000 impressions but only 4 clicks. Well, that pulls down my quality score and increases my overall costs. So, those terms aren’t working either. Clearly if phrase match does work for this term, exact match won’t either.

Here’s the crazy thing - I did have 19 conversions, all time, for this keyword “design school”, but because of automatic matching and all of the terms Google shows my client for - I simply have to turn off the term. I fold. Google wins. My client doesn’t.

I exported all the keywords that Google has come up with for “design school”. For this one term, they spent over $7k. I can clearly see nearly $2k in spend that has absolutely no relevancy whatsoever to “design school” simply due to automatic matching.

I’m asking Google Adwords for a refund. And, I’m just getting started on this path. I wonder how much more I’m going to find in non-converting or poorly converting keywords due to automatic matching.

I’ll report back as I learn more about why Google seems insistent on showing my client for non-related terms and what they can do to help me. And by the way, you just can’t add more negative matches when you’ve got 188 negative terms based on the list thus far.

Posted in Paid Search Tools, Search Analytics. No Comments »

 

Negative Matches for a PPC Conversion Audit

October 22nd, 2009

by Laura Thieme

Do you ever note the cost per conversion column in Google Adwords? Let’s say your CTR (click-thru rate) is good, and your conversion rate is good (CVR). But what about that cost per conversion column in Google Adwords?

Negative Matches

Above is an example of a PPC Conversion Audit where I’m working on lowering the cost per conversion. Google Adwords has made it much easier to see what terms people are clicking on, and not converting as you see from the screenshot above.

Example: 1% CTR - working on improving by working on keyword accuracy
Example: 4.25% CVR - not bad, but might still be even better when you look at negative match opportunities
Cost Per Conversion: $350 over the accepted amount per campaign conversion cost or acquisition cost

So, if you want to lower cost per conversion and increase conversion rates, look at the keyword that has:

1) The most clicks in a given period of time - where you have high cost of conversion and/or lower than desired conversion rates

Select keyword, and then select
2) Click on that one search term at a time - then click on See Search Terms in Adwords, click on Selected

Negative Match Term Opportunity

Once you’ve selected “see search terms, selected”, you might be able to see some keywords that are absolutely non-relevant to your product or service.

For example, I can see for “design school” the following keywords, that are costing my client about $11 a click and exceeding them in acceptable cost per conversion rates or cost per acquisition rates. Warning - the keyword is on broad match - but has had some conversions that would be difficult to have strategized. So, what we want to do is get rid of the non-relevant terms, or add to the negative keyword/match list.

1) design hoodies
2) designs for church bulletins
3) school embroidery design
4) school spirit designs
5) logo design
6) logo design sweater
7) school hat design
8) class ring design

Many more of these terms exist. What happens when your Adwords campaign has keywords on broad match? You have the opportunity to update your negative keyword list on a monthly basis, but do you? Google has made it much easier for you to see the terms that don’t convert, but receive clicks. The question is are you utilizing this every month to further reduce your cost per acquisition?

Within one month of turning off keywords that never converted in 13 months, getting rid of terms that had a very high cost of conversion, and enhancing our negative keyword match list, we have accomplished the following:

1) Increased click thru rate from .04 to 1.04%
2) Increased conversion rate from 2.75 to 3.61
3) Cost per conversion has improved but still needs more work; currently $400 over CPA

Caution:

One of the reasons that cost per acquisition can be so high is that the campaign could be new, and keywords need to be better researched to begin with.

Another reason could be content match

Another reason could be negative matches are lacking, or root negative terms aren’t added on monthly basis based on actual search data

Another reason could be turning on and off your campaign throughout the month to leverage your PPC budget

Another reason could be PPC conversion code is not installed correctly

Have Bizresearch pull a 13-month PPC Conversion Audit to determine how to increase:

1) CTR for terms that are more likely to convert
2) CVR for terms that have lower cost per conversion

Posted in Paid Search Tools. No Comments »