PPC on a Limited Budget
Following on from the How Much Cash For PPC Clinic review, Keith sent in a very similar question about the level of budget you should allow when testing out a new PPC campaign on a limited budget. He asks:
“Whenever I give PPC affiliate marketing a go, I always end up blowing my budget in a few weeks and find myself not having enough time to find those profitable keywords I need. When my budget is gone, I am a little scared to continue as I could lose a lot more money.
My question is, do you recommend having a large amount of capital to give it a go properly and if so, how much, €1k?, €5k?, more? … “
Answers:
Kieron
If you’re not making a success of PPC now, then throwing more money at it isn’t going to help. Forget about a budget, just spend what you can afford and nothing more.
Shane
I don’t think a bigger budget will help if you are strapped for time, you’ll just have more data to crunch and at more expense, although that may be the incentive you need to get stuck into the optimisation if it’s just the lack of capital at stake that’s meaning you don’t give it the attention it deserves.
It seems your main issue is time and not money, so perhaps make use of the day parting option and run for a few hours each day, or lower your daily budget right down so you gather data over a longer period, but only for several hours per day instead of all day.
This way you’ll get any weekend/payday traffic and more sales if it’s a product that converts better at weekends than in the week or vice versa. Then once you have time, analyse the data, run search query reports, add negatives, break adgroups down, write more targeted adcopy etc. and aim to improve your Traffic Quality (TQ) scores etc. to get cheaper traffic.
I’m not sure how you are measuring keywords to find out which ones work, but one thing you could do is pass a keyword in the url if it’s direct to merchant PPC (ideally I’d say pass a keyword ref id rather than the exact keyword). If you are doing traffic to your site then bouncing to merchants, you could use php to do the same, or at least use an out page and add google conversion tracking to it to record off site clicks as “sales” to track which keywords actually drive traffic into the site, and result in clicks leaving site to the merchant. It’s not ideal but it’s better than nothing, and at least you’ll see keywords which drive traffic to your site and result in no action.
Gathering data is the key regardless of budget; it’s as important to find out what doesn’t work as what does, so you should be tracking keywords as deeply as you can on every campaign.
Editor’s Note:
Don’t forget, if you’re running active PPC campaigns, why not take Shane up on his offer of a Free Adwords Healthcheck at www.freeadwordshealthcheck.co.uk.
hi, Keith, Steve, everyone,
here are my 10 quick tips for trying out PPC on a budget:
1. make sure the content network is turned off (the search network is usually more profitable & always easier to control).
2. use lots of negative keywords to avoid advertising to people who aren’t going to buy.
3. start off by trialing longer tail terms with buying intent. eg avoid things like ‘laptops’ or ‘sony laptops’, go for lots of variations of ‘buy sony vaio vgn-ns30z laptop’ instead.
4. use *only* exact match if you have a very small budget. open up to phrase match if that works.
5. if you can, ask the merchant which are their higher converting products.
6. use very tightly organised ad groups. make sure the keywords you’re targeting are included in the ads & that they all go through to totally relevant landing pages (eg. a product page for the product, rather than a general category page).
7. try to target terms with less ad competition.
8. if you’re controlling the landing pages (eg. you’re sending the clicker to your site then onto the merchant), make sure your quality score is as high as possible.
9. always use 2 or 3 different ads per ad group. turn off the ones that don’t work & try to improve on the ones that do.
10. just give it up if it’s not working. if you’re not hugely interested in learning ppc & you don’t have a lot of time, forget about it and concentrate more on the things you’re doing well.
hope they’re useful.
dan
Hi Dan.
Cheers for some excellent tips there.
When trying long tail keywords such as “buy sony vaio vgn-ns30z”, how extensive would you make the negative keyword list? Just the other Sony models or other manufacturers as well?
hi, Steve, how are you doing?
warning: this will probably get seriously boring!
With that particular keyphrase ‘buy sony vaio xxx’ on exact match you wouldn’t have to worry about negatives as it has clear buying intent & there’s (because it’s exact match) there’s no risk of hitting unwanted searches.
But with something like “sony vaio vgn-ns30z laptop” on phrase match (ie. there is rough buying intent because they’re searching fairly specifically, but there’s a chance there will be other terms in the search) you might match against things like “broken sony vaio vgn-ns30z”, “sony vaio vgn-ns30z manual”, “sony vaio vgn-ns30z photos”, etc. I’d add as many negatives as I could think of & would use a couple of extra tools to help find them.
Maybe this will help – let’s say I was actually going to trial advertising sony laptops & wanted to keep risk really low, here’s a process I might go through:
1. Set up a campaign “Sony Laptops”
2. Use google’s keyword suggestions tool, twitter search, google blog search, etc to figure out which are the hotter models.
3. Set up exact match ad groups for a few specific models. For each model I might set up a couple of ad groups initially: One targeting ‘vaio modelname’, one targeting ‘sony modelname’ – that way I can get those keywords into the ad nicely & can track them separately a bit easier.
4. Once I have some actual traffic data & can see how much volume there is, I would duplicate the better campaigns into phrase match and maybe set up separate groups if things like ‘uk’ or ‘buy’ (or any other modifiers) are popular.
[Now because I’ve expanded into phrase match, there’s a risk i’m going to hit searches where the searcher definitely doesn’t intend to buy.]
5. Look for words that would indicate the searcher isn’t looking to buy, or isn’t looking to buy the laptop itself. Examples might be: broken, problem, free, photos, manual, discount, used, secondhand, support, accessories, batteries, mouse, etc. I’d add these in at the campaign level so I don’t have to duplicate work & can keep track of them easier.
6. If I’m sending people to my own site (rather than direct to the merchant), I’d keep an eye on the actual search phrases being used & add any extra negative keywords I find into the adwords campaign.
7. Continue to expand the stuff that’s working & ditch the stuff that isn’t.
Tools I might use to speed this all up would be:
. Aaron Wall’s amazing keyword list generator: http://tools.seobook.com/keyword-list/generator.php
. A google analytics hack for viewing the exact search queries.
. Google’s keyword suggestion tool & the sktool.
. The Google AdWords editor
Any use?
dan
Hi Dan.
Everything’s going well thanks, and that wasn’t boring in the slightest!
It’s helpful to hear exactly how someone with the experience in PPC would go about doing it, as a lot of people seem to burn through their cash without thinking about it properly.
I’ve yet to delve into PPC, so it’s definitely helpful and I’m really enjoying these posts.
Incidentally, I also found a great post from Mark Boyd (@hairycornflakes) on his experiences with negative keywords at:
http://www.markboyd.co.uk/2009/06/negative-keywords-not-so-negative
Cheers.
Thanks for the replies guys, oh and Dan
Fantastic advice!
@Shane, “Gathering data is the key regardless of budget” – Noted, I use ‘encrypted’ sub ids so I can see what converts best.
Cheers
Keith
Hi,
Thanks for a good site,
I have a side question, I have started running quite a few PPC campaigns though google and so far I have been using only one account – Some people have advised that I use a account for each of my campaigns as it should work better.
Can you comment on this as I run more than 40 campaigns on and off and having an account for each of them would mean that I have to put money in 40 accounts instead of one and I would not have a quick overview as I have now.
Thanks
Nils
Hi Nils.
Cheers and thanks for your question!
I’ll refer this to one of the others.
Hi Nils, I can’t see any benefit from opening more than 1 account. Besides it’s against Google Adword’s terms and conditions, see http://adwords.google.com/support/aw/bin/answer.py?hl=en&answer=8759
I think you’ve been given a bit of duff advice there Nils! As far as I know, Google likes you to operate from a single account to prevent such naughty behaviour as double serving. I’m sure they have given people with massive accounts permission for a second one, but I really don’t recommend 40 different accounts.
Hi Again
Thanks for your reply,
After I read your reply I have found something that might show something different;
I run a few mobile campaigns where I am allowed to use the merchants URL and I checked a few keywords and some are not showing on all campaigns because of this message according to google:
———-
This particular search term is already associated with an ad from one of your other campaigns or ad groups. Only one ad per advertiser may appear on a given page per search term. When two or more of your ads are eligible for a single search term, we display the one with the highest ranking at the moment.
———-
The quality score is 10/10 and it is a good term for companies with cheap mobiles and as an affiliet it would be great if I could be shown for the same search term as it gives me a greater chance of converting.
It is different companies and they are competitors, but I guess I am double serving even though it is different companies and different URL’s. They have been setup with each their own campaign and ad groups.
What do you think of this?
Hey Nils, it would still be considered double serving. If you were a paid search company what you’d do would be to open one account per client and then manage them via a Google Adwords management account which allows such companys to service multiple clients.
However as an affiliate, if you double serve in such a way you will be in contravention of their terms and conditions. If you get caught, they’ll shut down all your accounts and issue you with a lifetime Google ban. The best thing to do is work out which merchant is best to send particular traffic to.
Hi Kirsty & Kieron
Thanks a lot for your help and answers
Regards
Nils
[…] recently asked the Affiliate doctors this old chestnut: “Posts about PPC on Limited Budget and How Much Cash for PPC are a fantastic resources for newbie “PPC Am’ers”, with a lot […]