After our dismal experiment with trying to track impressions, Rand's course of action was to have me dive in headfirst and manage our PPC campaign for our Premium memberships. I half-assed the campaigns in August '07 and retooled them in November, when we redesigned our landing page and added a few targeted pages. The clickthrough rates ranged from "need improvement" to "pretty good," with our best campaign averaging around 3.5%. Our conversion rate to our "Choose" page was great, ranging from 75-100%. And then I paused our ads.
"Why?" you ask. After all, our clickthrough rates were decent (though they could be improved) and our conversion rates were astounding. Why on earth would we stop running our campaigns?
Because our conversion rate of those who went from the "Choose" page to our "Completion" page is terrible. In AdWords we track conversion rates from our ad to our "Choose" page, which is where users get to choose which membership they'd like to sign up for. However, with our analytics program we track those who actually made a purchase, and those numbers tell a different story. At far lower than a 1% conversion rate, it was time to assess what we were doing wrong.
For us, it was a no-brainer--our signup process is awful. We lose a lot of people in that process, and Rand himself couldn't even sign up for an account without experiencing issues. Our devs are in the process of rolling out a new payment system that will make signups much smoother and easier, and that should be deployed at the end of the month. Faced with that knowledge, I had a decision. Do I let our ads run and continue to lose money for the month of February because of our abysmal payment process, or do I pause our ads, wait for the new system to be implemented, and resume them so I can test and see if the change improves our campaigns?
I really learned an important lesson here, and I wanted to share this lesson with other PPC noobs like myself. My first encounter with AdWords, when we were running our impressions test, was pretty embarrassing. I thought you could just set up a campaign and let it do all the work for you, but that's certainly not the case. You need to check your campaigns constantly, and you certainly don't want to be losing money in your advertising attempts. In my experience, it's imperative to know when to hold 'em and when to fold 'em, so to speak. If something's wrong with your campaign, sometimes the right thing to do is pause your ads and figure out what's causing the problem, then make those changes and resume for testing. There's no reason why you should have to dump unnecessary amounts of money on your campaigns, and we've now learned that.
By: Rebecca Kelley
When You Should Pause Your PPC Campaign
Paid Search Marketing (PPC)
The author's views are entirely his or her own (excluding the unlikely event of hypnosis) and may not always reflect the views of Moz.
Good call, Rebecca.
Doing PPC without actively doing analytics is exactly like running your business without any bookkeeping: sure you can do it for a while, and it seems easy, but eventually you find out you are losing money the hard way!
I'm an SEO guy, but I've been doing a lot of PPC lately. Remind me to tell you about how I had to take out a second mortgage on my own home due to a PPC issue with a client once!
Here are some other reasons to stop a campaign:
It's TOO successful - at a certain point, most companies can't handle the volume and need to slow down and focus on their current clients. I also had a pharmacy client once who received a phone call from Canada Health regarding them selling all our Avian Flu medicine stockpiles to our friends south of the border. Oops, my bad...
It's got a lousy ROI - obvious, of course.
The PPC vendor (Google, Yahoo, etc) is unable or unwilling to fix an issue that prevents you from properly analyzing your date. Yahoo is infamous for their reporting center going down. When they stop reporting, I stop campaigns, unless I have direct orders not to and my client knows the risks.
It's the wrong time of the day. This is more of a dayparting issue. But stopping or slowing down a campaign at certain times (weekends, late at night, etc) is sometimes necessary and prudent.
$$$ - If you are concerned about running out of funds before the end of an official campaign date, you may have to slow down or pause things. It's better to do it earlier in a controlled fashion than later during an emergency, or Google making your campaign go dark and you not finding out about it until you check your email a few hours later - ouch. That's lots of fun to explain to a client if it's their credit card you were using. Even worse it it's yours.
Crappy service. We once had horrible service from Google Canada. Right during the end of the year, they all shut down and went on holidays (sorry, be back in January). Hasty phone calls to Google HO only got us "we aren't allowed to help you, sorry". We now work with Google US instead.
Click fraud. If you are getting a lot of click fraud, you may need to either shut off or severely limit your content network campaigns until you come up with a different plan. Or abandon them all together.
Events. Sometimes, certain events may drastically change your business - 9/11, as an example. Certain holidays or political events too.
Death. Yup, a friend of mine had a PPC campaign once and the client died. It wasn't until the credit card stopped working a few months later that he realized it. Oops.
Content. You may find that some PPC vendors don't/won't carry certain content (guns, etc) and if you change your business model or product line you need to know this and act accordingly.
Stupid competitors. Every once in a while you may find that you get some new competitors that bid based on ego rather than ROI. Occasionally if you get more than a couple of them due to the latest "hot fad" hitting your niche, the bid prices will skyrocket and you may be better off going dark or on idle until these fools spend themselves into oblivion and the bid prices adjust back down.
That's what I can think of off the top of my head at midnight, so I'm sure there is more.
People are beginning to learn that PPC can be a wonderful thing, but, as you've discovered, PPC is like a loaded gun - powerful and easy to use, but if you aren't mature and careful where you aim, you can hurt yourself or those you were trying to help.
Ian
Ian, I'd love to ping you when I resume the campaigns. Maybe you can give me some pointers.
No problem! I don't fancy myself an expert, but I do have a large (and continually growing) collection of mistakes I can share - which sometimes lets me fake expertise... ;)
Ian
@ mcanerin: some really good points. As the one who has very poor PPC experience, I really picked up something new from your comment... [I am not sure why but if I were to choose, I'd prefer the first reason (as the lesser evil).]
As per my experiece, we usually stopped the campaign when the ROI was so low that it was no more worth the effort... Internal optimization never helped... I suspect that was due to the niche - in entertainment niche people either want to buy (and then they are ready to bear any inconveniences) or just never meant to buy (and came just out of curiosity)... but again I have to little experience to judge... With SEO niche (I was trying to advertise a link building company) I never even managed to get people to click the add and paused the campaign until I figure out what to do ...
Some great points: I'd like to second Time of the day. I've seen timing have huge effects on PPC campaigns. In lieu of shutting off a campaign completely, restricting it to your highest converting days and to daytime hours can work wonders. Google even lets you tweak your bids during different hours if you want (you can run your ads from 10am-5pm for example, but set them to 150% of the base bid from 1-3pm).
Another trick I've found useful is to set a low budget for high-impression keywords but then split off your long-tail (assuming they're higher performing) keywords into their own campaign(s) so that they don't run out of budget. Otherwise, your budget hogs soak up the money and your big performers only get a partial day to run.
@mcanerin: I'd love to see you expand your comment into a YOUmoz post on PPC tactics.
@mcanerin: "Remind me to tell you about how I had to take out a second mortgage on my own home due to a PPC issue with a client once!"
I want to hear this story too!
Rebecca - I have lots of PPC experience and I think I can offer one piece of advice: High Click-Through rates is not the key to success -- only high conversion rates. In fact you may be focusing too much on getting people to click your ads who are not great candidates for premium membership and thus hurting your conversion rate. You may want to get only a quarter of the same market to click your ad (the quarter that converts).
If my goal is to sell something then I make that abundantly clear in the PPC ad. There is no point in getting click-through from people who are only interested in free. If I were you I would create ad text like: "Premium Memberships starting at $49.95/month" or something that clearly lets people know what you are selling and what the price range is. The people who choose to click on that ad are VERY likely to buy. Your Clickthrough rate will go way down, but the people you lose are not your target customer anyway and you've saved money on false impressions/clicks.
I know that high CTRs aren't necessarily a good measure. I'm still learning how to tweak our ads, though, so I figure I can work with variations and continue to play around with them. I'll try a group that touts our memberships in a clear and concise manner, and I'll see how they perform.
Thats a very good point - that way although you will definatly drop your clickthrough, you will also reduce waste spending on idle surfers and geta more reasonable ROI.
I had to pause a campaign due to a domain expiring. Unfortunately, I didn't know it expired until it was too late, since I didn't register the domain. It was a costly lesson. The good news, I still have them as a client.
I've run a few PPC campaigns, and the potential problems with them are a nightmare - especially when you're inheriting someone else's crummy design/development.
I have a few PPC clients right now, and I'm happy for the experience since I've primarily been concerned with SEO...and wow, the issues abound. It's difficult to definitively declare where my duties as the manager of the AdWords account end. I've found myself leaking into business/sales management for one company, asking them questions like "can you handle this many new sales, or is there going to be a bottle neck?" or "have you looked at adjusting your pricing - if you cut into your profit margin slightly you'll make it up and then some in volume."
There is just a lot going on once you're involved in not just driving traffic to the website but optimizing the flow of that traffic through to the point of conversion. Everything is on the table - the product, the brand, the reputation, usability, price point, etc. What I've realized is that quite often what clients need isn't just a PPC manager but a marketing/sales manager as well.
I couldn't agree more. PPC management forces you to look at the whole business and product and fine tune more than just the ads themselves. Markets are extremely price sensitive. Customer service is essential in many markets. Business processes must be fine-tuned. Landing pages need to be changed and tested and retested. Adjustments must be made as new products are introduced and best-sellers are backordered. And you could keep working it to infinity.
I have very little PPC experience, and the whole thing about them being almost a full-time job is spot on.
I went into it thinking it was a set it and forget it kind of thing - and found out it wasn't.
You did the right thing, if the sign-up process is broken - everything else in the sales process would be tainted.
The ad companies love if you set it up and forget it (especially if you're on big fat credit card!)
What you've come away with here is key: your checkout needs to be ROCK SOLID. Fix the dev issues and make the selection process a more seamless function of the purchase.
Finally, if you can't get a purchase, try to get an email sign up so you can send a Premium blurb every now and then and entice towards a future membership.
I'm finding that it does require more attention that I initially thought, so I'm glad you feel the same way!
Rebecca,
I admire you for sharing not only in SEOmoz's successes, but also its, let's say - "challenges".
First, I think it tells everyone that just because SEOmoz is expert at SEO, doesn't mean there aren't other areas in the customer acquisition process in which it can't improve.
Secondly, even with a solid landing page or pages, most efforts are only as strong as their weakest link and in this world of impatience and "quick clicks", you need to have almost flawless execution to achieve the most desired result.
That said, the most immediate thought that came to mind was my initial reaction when, as a month to month premium member, I wanted to upgrade my account to a six month or one year membership and couldn't find a quick path from my profile page to do so. Unless I missed something, it required an exchange of emails - which is cumbersome for a simple Paypal or credit card transaction.
So, in implementing this new process, I would also urge you to take a look at the membership upgrade situation from the existing profiel page and to ensure that there is an easily visible button and process to handle this transaction.
Thanks.
Our upgrade process should improve with the new payment system. Damn you, Paypal!
I hear you on the sign up process. Dude, it's awful! Glad it's in the pipeline for a fix.
running ppc is definitely a full time job. we have someone who basically works on ppc all day everday.
we also have clients who ask to have their campaigns paused for various reasons. as mcanerin suggests, one client in the mortage biz gets too many conversions at times and asks us to pause the campaign sometimes because they cannot handle the volume.
another client sells products on their site but does not have a shopping cart, customers have to call to order (shopping cart coming soon!) so we turn it down during the hours that they are not at the phones.
i'm just thankful we have someone to take care of all this. i'm adwords certified but it's definitely not my favorite part of sem. good luck rebecca.
Hi Rebecca,
Nice to see some paid search posts. :)
Do/Did you know exactly what ROI you were gettingfrom AdWords? If it was positive there is absolutely no need to pause the campaign. You mentioned your goal completion rate was less than 1% but not if that was still profitable.
Adwords is no longer a set and forget get rich quick tool. It takes a LOT of work and sleepless nights to make it work, but when you do, it really, really, pays off.
It was definitely a negative ROI, which is why I decided to pause the campaign. I'm hoping that by implementing our new payment system, things will drastically improve.
Gotta agree with you on that one Dave!
Yep, it's a rude awakening to realize how small of a % of strangers convert. Everybody goes through it. The #s will be extremely low when people aren't familiar with your services/products. Gotta lowball bids on tons of kws, segregate ad groups & relevant keywords like crazy, then make constant adjustments once you get an idea of what's providing positive ROI.
My advice would be to at least bid on your "brand names" keywords - seomoz, seo moz, rand fishkin, etc. When you send people who are familiar with your brand directly to a landing page with a targeted ad, it should result in a kickass ROI.
Brand names are just the low-hanging fruit. It takes a lot more fine-tuning for generic terms. Have you tried an ad group targeting terms like "SEO tools"? With some changes to the premium landing page, those types of queries seem promising.
Of course, I've never done PPC for an industry like SEO. I imagine it's no walk in the park.
We actually were bidding on our brand name, and we had zero impressions and clickthroughs reported. Also, I did set up a separate campaign for "SEO tools" and related keywords. That campaign, along with an articles/guides-related campaign, were both doing well.
Rebecca,
How much of this was actually due to the term you were using and the target audience? You were bidding on your brand name (always a good idea) but the sort of people likely to use that as a search term would be above average in search engine usage and knowledge.
So, and I'd be very interested in other's input, could it be a case that your target audience are so search engine savvy as to not click on your PPC ad, but on your organic listing instead?
For our financial services brand name we get about a 50% split on picks between PPC and organic, but then we're not targetting search professionals.
i agree with dave and smartSEO. if you are still retaining a positive ROI, the very last option should be to pause the campaign since you can always lower your bids. and it's always best to compete for your own brand on ppc bids, if only so your competitors aren't stealing your traffic. great tips, guys! and vingold is right, ppc is full-time for many individuals, which is why some SEO companies i know of have dropped their ppc services :D
hope everything goes well with your signup/checkout system.
As I stated above, we had a negative ROI, but you're right that if you're retaining a positive one, you shouldn't necessarily need to pause your campaign.
yes, good call on pausing your acct, i was just agreeing with dave's point.
One of the hardest parts about PPC ROI is that you can have solid numbers, theoretically, and still lousy ROI just due to the dynamics of your niche.
Case in point: I've got a client with good CTR (6-8%) on some ads and solid conversion (2-4%), and they're still losing money. Not only are bids a bit competitive (and probably too high) in their industry, but they're effectively a reseller, so their margins are relatively thin. Even if we squeak a bit more conversion out, the numbers may just not be in their favor.
@msdanielle: Hmmm... That was actually 'SharpSEO', not 'SmartSEO' :)
Good one Rebecca, I am dealing with this exact thing right now on my emloyer's PPC campaigns. The checkout process needs improving.
I wish I had the entire day to dedicate to messing with PPC optimization but with all the other things to do around here, there's never enought time in a day.
Just a week or two ago I signed up for your premium membership and I did think it was a bit weird but I figure it out.
Thanks and good luck with it :)
That's the exact same problem I have--so much to learn, so little time to learn it! I'm kind of learning as I go, which may not be ideal because of the ad spending, but it's working little by little.
as long as your communicate this (which you have) with your employer. they understand and take the amount it cost as a ''education fee''...
.rb
Nice post and glad you're fixing the sign-up process. I also got caught in the monthly upgrade to yearly conversion problem. I also agree about the Paypal subscription services.
One item I didn't see in these posts for stopping a PPC campaign is if the web server is down. I use a site monitoring service (Pingdom) and if the server goes down, I call the company. If the outage looks like it wont be fixed in 30 minutes, I pause the ad campaigns. I turn the campaigns back on when the server goes back up.
I've found that when tracking conversions some of our lowest CTR keywords proved most profitable. Why? We are targeting a very specific searcher who doesn't type in a very specific search. So instead we target the more generic terms but write our ads in such a way to appeal to our target but nobody else.
So we have a successful campaign with a very low quality score (and higher per click costs) but overall some of the lowest conversion costs of all our campaigns.
Sometimes searchers just aren't that sophisticated. So in that case, this strategy may work well for your campaigns. I don't think this applies to seomoz's target market but you never really know until you test...
This is a reminder that all serious websites have to deliver end to end: from their marketing, PPC, SEO and content through to visitor conversion by having usable functionality and easily navigable paths. Page load times and host reliability - all of these things have to be optimised to provide the best ROI. It's not enough being great at just one or two of these anymore, - running a successful website is a pretty complicated juggling act. Good discussion.
Odd that I wrote about PPC and landing page optimization today and then stumbled upon this post in my SearchCap newsletter.
I think the true lesson you learned is that even if your PPC campaign looks good, other things can influence its end performance in conversion. The influence can be so great it doesn't matter how well your PPC campaign is put together. Even the most optimal PPC campaign will never convert a user going to a very bad funnel (and the person managing the PPC campaign might think they're just doing a bad job!)
Good work realizing that your PPC spend was inefficient though and being willing to stop it. Keeping track of all the factors that make PPC successful is really difficult and takes a lot of managing.
Since your pain point is your funnel though, the obvious next step would be to fix it. Since your PPC sounds fairly successful right up to conversion, I would take everything you learned from your PPC and apply it to your whole funnel. What messaging was working there? What offers or features got you clicks and how can you emphasize those things throughout your funnel? Carry those winning value propositions on every page and make your visitors feel comfortable and excited about trading their money and personal information for your tools and expertise.
Good luck! It's tough doing these projects, but they're always satisfying when you can show how much improvement you made.
-Billy
It's amazing how incredibly complicated it is for an average person to set up a successful campaign. The fact is, unless you are sophisticated PPC marketer, there are too many ways to mess up your campaign.
Specifically, structuring your campaigns correctly, bidding properly, split testing effectively, choosing an intense and comprehensive list of keywords, etc.
The complexity and amount of time needed to manage a successful PPC campaign is only going to increase as competition increases in the months and years ahead!
Wow Rebecca,
Great post. I love to see the masses speak their mind. It's like a MENSA meeting of online marketers.LOL.
I have a few bits to add, I hear alot about "individual" account management (adwords). How many words are we talking about? I also keep hearing the same cry " I have no expierience ", well the percentage of industry professionals is rapidly growing and there isn't any real qualifications to validate one reply from the next.
With that said, I manage around 50 individual businesses PPC campaigns, I use between 1500-2000 words terms and phrases, across 98% of the net's search engines and major news services. I know this is going to sound "pitchy" but I use a software program that tracks the results of the keywaords ( ie word to - to a phone call or email)so far this sounds cookie cutter, but this is where it gets good the list of words terms and phrases is continuosly optimised (add words that work-delete low performance terms). Lets add a handful of behaviorial targeting and now we can have a campiagn with high ROI and low CPC. The system we use bids according to the performance of the keyword (i.e. word=phone call bid to 1st position..word is used for due diligence= bid to bottom of first page or top of 2nd) With that formula I can blanket more words and search engines and control cost-per-click. With all that new time I have saved tracking individual word or phrase results I can then focus on the behavior of my customer and focus on the conversion. I could go on for days about the success of a results-oriented-bid-management-system, or you could give on ea try for yourself. I use WebTracker from Higher Images. I would love to have 1 company to worry about and 1 focus of my energy's.The moral of this comment is to keep your head up, use technology to work smart and not hard, and finally take each opinion and bit of advice and find the hidden pearl of wisdom. Our industry has alot of chiefs and in order to stand out, you don't have to be the smartest you just have to prove ROI. I write alot about similar topics: https://www.higherimagesblog.com
i find one of the biggest challenges of ppc is setting and forgetting and not just going in and changing things after hour two or day one.
from my experience it usually takes some time for things to make it through the system before get some solid actionable results. on more than one occassion have seen results totally change and stabilize by letting a campaign run for an extra day or two. would've lost out on opportunity if i would've acted immediately.
with that being said if the campaign is totally tanking then taking it offline and optimizing makes complete and total sense.
much like everything else in the online world, things are totally different depending on the industry, audience, etc.
Also keep in mind that you can get very granular - one of the upsides of PPC - and only pause specific _keywords_. So if you have, say, more informational terms you want SEOmoz to continue to be found for, which are not as signup/conversion specific, then allow those keywords to run as part of your "brochure" campaign.
Then simply pause the conversion to paying subscriber-related keywords or campaigns, until that process is fixed (and yeah, count me amongst the challenged by the signup process! ).
You can easily separate overall branding from revenue-generating-specific areas in this manner.
thanks for sharing your experience, you did the right job when stopping the PPC and enhancing the slowest chemical reaction... the registration!
I manage PPC campaigns for some clients. It takes work and creativity and late nights crawling through lots of data.
It's so important for clients to understand the value of the landing page quality.
It does two things that are important:
1) It lowers your PPC if you get a "great" quality score on an ad.
2) It converts customers who want to sign up in the first place.
Ah, the struggles of Google AdWords and the befuddled orphan landing page. ;-) Pause for Peace...
i am a seo jus hav't had any experience about PPC and recently start reading about ppc and get confused about different way of advertising especially pay per click, Page impression, and CPA.. bannner and text etc could you guys can tell me which on most cost effective in term of money, time and quality click as i am about to start a ppc for one of my client thanks rebeeca i been following you articles from quite long now.. i appriceate your helps guys
I'd focus on one only and familiarize yourself with it. For example, you could start a PPC campaign and work with a small budget to get a feel for how it works and to learn more about it. You're going to spend money, but hopefully you won't bleed too much in the learning process.
thanks i set up now round about 25 campaign for my client but we still waiting for website, as there are some bit and but left to finalize the final project. mean while i ahd some more time to learn about PPC. i am thinking to start with very cheap click and then i will move on from there...by the way how you think about CPA... i donot know how to set up, to me its look very interesting i think it will help me very much..
here is a very useful article about how social Media work
Rebecca - good to see some seriously honest posting, most companies wouldnt put something like that up.
I would like to add my 2 cents (I manage and advise on various PPC campaigns)
Looking at your business model of conversion and being a free member for a while now (except for that brief stint with the free premium membership you offered) - I think looking at direct conversions from PPC would be not the right thing to do.
But even before I go there, do you drop a cookie that is trackable by more than 30 days? some conversion takes time and PPC is no different - revisiting first point of entry and end conversion may show a different picture.
Also, what percentage visitors became free members? I am glad that SEOmoz doesnt actively email us to convert to premium members, but honestly I think this should be revisited. After all, you have driven targetted traffic to the site using both PPC and SEO, got a sign up and first commitment, so why not press the advantage? Relying on one channel to convert isnt totally realistic, PPC works hand in had with many other channels, and follow up to PPC visitors may turn the conversion tables...
www.MarketingExperiments.com - show that to whomever creates your landing pages.
You are using multiple landing pages correct? And I'm not talking about A/B testing, I'm talking about having the value proposition of the landing page match the value proposition of the adwords ad. If your ad says SEO Tools, the VP of the landing page needs to be SEO Tools. If your ad says SEO Articles & Videos your VP of the landing page needs to be articles & videos.
Yep, we've got three, and I wanted to add a couple other specialized ones.
I guess I'll jump in and comment about PPC. One thing that has always confused me, is that if you have a brand recognition website, such as seomoz.org, where people tend to think of it as an authority, and anyone that is in the "business' of SEM, SEO, etc, knows about it, why pay for getting visitors?
The alternative is that if people don't know who seomoz.org is, are just arriving on the "scene", what are the chances of them converting right out the gate?
how about bidding for content, - get a newbie reading... then joining.
the bids for blog articles should be cheap enough ...etc.. (perhaps thats a horrible idea I dont know) think I heard it from aaron from seobook but he was advertising viral content to get it started.
.rb
SEOmoz's chances of converting a "stranger" may be pretty low, but they're model is also one where the marginal profits on premium memberships (once they've paid expenses) are very high. So, if they have the 1,500 members that cover their costs, to pick a number out of thin air, the 1,501st buyer is 95% profit. That gives them a lot of play on additional marketing channels like PPC (as long as those channels aren't cannibalizing their other marketing efforts).
This is what I refer to as "entrapment". You pull people in via PPC. low pressure tactic and offer an excellent free service (free mebership, excellent info, regular updated blog, plenty of community participation).
Over time the free user builds a relationship and trust with the brand and would convert into a paying member to access all the premium goodies.
Its a great model if you dont mind a long purchase cycle.
it is a good thing you are talking about that. Actually I use Adwords for different accounts, and I also do use Analytics.
I order to improve my PPC campaigns, I used to get the figures from Adwords, and tweak my landing pages, change from large to expression, my ads, and so on...
Going further down with Analytics, I found out that if you go to
Traffic Sources>Adwords>Adwords Campaigns
and you then click on the campaign you want to have a closer look at, you have actually the Bounce rate that appears as well for each keyword you are bidding on.
So, on one hand I have the CTR on Adwords telling me when people do click on my keywords, but even more useful, I know that once they clicked on those words, I have x percent of bounce rate per keyword. Which means, that either the landing page is not the right one, or that the people are not seeing the information they are looking for on that page.
Since I want to improve the conversion rate for my clients, and make the user experience the most profitable for them as well, I think that the combination of the 2 software and analysis can make you achieve those goals. Further more (and that was the bottom point of your post Rebecca), when people are clicking on the PPC ads, we have now a lower Bounce Rate, therefore we are losing less money !
I believe Google has a page optimiser function for your landing page that plugs into adwords and Google analytics. I think it involves designing some page variants and Google will give you lots of yummy data on which is converting properly. Of course if you've got a 2-step conversion transaction it may not be so easy.
Be very careful about pausing non-performing keywords too... the more generic ones are often the driving force behind customers coming back a week later to use specific variants.
e.g.
"Laptops" may not convert to a positive ROI but if you pause it and notice a considerable drop in "Sony Laptops" or even "Sony Vaio123 Laptops" then you may have discovered some importance in the keyword weight of the generic term.
I'm not saying don't pause it - just watch the associated keyword ads very carefully in days following to make sure they aren't affected.
Your problem was really big (if you see from bird eye) and its always advisable to pause campaign in this situation.
Very smart to pull the ads. If it's broke, reload and retry.
I will say it is much more fun to pause a program because it is working too well. ;)
Thanks Rebecca, really interesting. I have never run a PPC campaigns so I don't have experience but I know one of the advantages (comparing with other traditional advertising) is that you can monitor almost on real time. This gives you the power of testing, analysis and also the possibility to make the changes you need to improve and start over. So it was a smart decision.
I've been doing PPC/SEO for some years now and while it can be a fulltime job there are a few tools that can soothe the pain. The most useful being the AdWords Editor.
It sounds like you should redo your conversion tracking. Your "Choose" page is not really a conversion event only a key page. You should track that page in analytics as part of your conversion funnel but not as a conversion event, only your "thank you" or "payment success" page should have the conversion tracking script.
I posted this on my site about PPC tracking. https://www.theppcguru.com/category/ppc-tutorials/ppc-essentials/
Also make sure you keep your quality score high as well. I am the SEM dept manager at an agency and I have seen time and time again two different clients both competing for the same keywords and the one with the highest quality score will pay at least one dollar cheaper than the client with the poor or ok score for a higher position (not that the highest position is best).
Also (I can add hundreds of these alsos) always do multivariant test before you launch landing page changes and always have three ads per ad group no more no less. This way your ads in the group is competing against each other. initially go into your account setting and distribute the ads evenly, then once you have an idea of which is performing the best switch it to "show best performing ads most often." every few days go in and pause the least performing ad and write a new one.
3% CTR is not bad but consider this:
1. the higher your CTR is the Higher your quality score (there are other factors too) will be and the less you will pay per click
2. 3% CTR is about average, when you use the technique i've been using your AVG CTR always goes up, never down
3. It may take a while but you can get double digit CTRs. I have gotten as high as a 32% CTR using this technique over time. The only way to achieve that sort of CTR with out luck that I know of.
The same is true with multivariant testing (I have several accounts we get a 19% conversion rate). There is a point of deminishing returns however. When you start to slow down with CTR or Conversion rates and your in the double digits start testing other adgroups/landing pages.
We actually track our "Choose" page in Adwords so that we can keep our QS high. We still track our payment completion page in our analytics, so we're able to compare the two figures and get an idea of how much abandonment we're getting.
My tests has never shown that tracking of key pages affect QS.
Here is what we have found affect QS:
1: CTR
2. keyword relevancy to ad text (particularly headline) and landing page
3. time spent on landing page
4. bounce rate
5. conversion rate
There are tactics/techniques for adressing each one of these individually. Too many to list here but feel free to ask any question you may have.
I have also tested the effects of the Title & Meta Description and found them to be very important for QS. I do agree that conversion tracking doesn't have much of an effect on QS. It may be a small portion of the QS, but I can't isolate it as having any effect.
Whip out Rand's SEOMOZ Black Card and bid on the word "mortgage" i heard that it converts really well for SEO sites.
You were absolutely right to pause your campaign. Something I really like about PPC is that it can alert you to problems with your site that you might not have considered previously.
For example, in a previous role I looked after PPC for a jobsite - I conducted a test to see which landing pages converted better for jobseekers - the generic home page, or a more tailored jobseeker landing page.
Conventional wisdom would suggest that the one tailored for jobseekers would convert better, but strangely enough - it didn't.
It was only once we'd seen these results that we all sat down and really looked at the jobseeker's page. It was a proper schoolboy error - the sign up (what we were counting as a conversion) was appearing below the fold. On the homepage the sign up was above the fold.
Basic enough error, but I think that if we hadn't have undertaken the PPC activity it would have been a long time before we noticed it!
Make your life easier - add the SE tracking pixel to your landing page. Remove later when you have tons of sales (if you want).