Product Price Searchengine - Practice and Ideas (14)
Hi guys,
we developed a price search engine like billiger.de / idealo.de / Shopzilla for six months and tested adwords for 1 month now.
Everything is ready to go, deals are signed, we earn money with every click on an item on our page.
Focused on DACH.
The problem
We planned to have google adwords as main traffic source. Ad-Quality Score is perfect 7+/10 but site quality is below average.
Thus our cpc's we are paying are way to high and we dont get a single keyword roi positive. We tried broad match, exact match, created ads for each longtail keyword....
Also we tried to improve the search relevance on our site.
Tips / ideas?
Does some of you guys have any ideas how to master google adwords or other sources for this? We also tried pops on keyword matching from rtx but also this didnt worked even if we target on ebay.de + keyword in combination.
We are open to any new ways / ideas, also to cooperate with some of you guys / building extensions / search solutions ... There would be just these requirements:
- no shady stuff
- in every case the user needs to click on the item on our page by himself
We won't post the url in public. If you got any ideas or questions let me know, i would love to find a working niche with you guys on this 
cheers
cash4nuts
It seems to me that Adwords is simply to expensive for your price search engine.
And it will get worse day by day.
To give you an idea; 5 years ago I was successful in a niche where the average click-price was $10, but today that is above $30/click,
and there is no way I can get it profitable anymore.
There are a lot of treads here at STM about facebook-marketing, I think that is a better option for your project.
Have you test mobile vs desktop traffic from adwords? We've dabbled in this sort of arbitrage and found mobile always works better. Are you tracking click-outs as conversions? Using CPA bidding?
Originally Posted by johnnyx
also we tried facebook but this seems a bid to broad to us.
@alexdmark
we see a big difference here but depends on the keywords - tablets works best for us atm. We're using clickout converions and also tried CPA Bidding - we will give it another try with more longtail keywords now.
What category/ niche are you focused on towards price?
How have you structured your keywords together in groups?
What types of keywords have you focused upon (generic, product brand, competitor+ brands or specific product models)?
You mentioned you are using broad match, what about broad modified?
How tightly themed is your copy to your keyword groups?
It tends to be the higher the ad ctr the lower your cpc can get, which is in relation to heavy copy testing of ad angles of your product focus and what your customers are looking for. This includes giving it enough time and data to make decisions upon.
Have you checked your search term report to see if you are showing up for relevant terms, which you can also be used to add to keyword negatives if the search terms are irrelevant or too expensive?
Are you targeting all devices (Desktop, mobile & tablet)?
Maybe better to start off in 1 device to get it working first then work outwards.
I'm not sure if Amazon still offers ppc and your products fits within the amazon space, maybe it's worth a look at especially as it's using Amazon's brand recognition and environment.
Are you using Adwords display network or search only, or remarketing?
You mentioned you did PPV targeting ebay.de, were you focused on the categories/brands/models or just top level domain ebay targeting?
Hi there, think you need run some time with negative roi - you need to build brand - people will going for price compare direct to you, not from google. Build email list, etc.
Hi cash4nuts,
when Adwords to Amazon arbitrage was still allowed, we made good money with massive longtail lists and very low bids, but i don't know if that is still possible nowadays.
If you need more details feel free to PM me.
Originally Posted by egliptor
We tried many types of campaign structures, also niche + keywords. Do you think google quality ranking also depends on structure?
In the meantime we tried broad / broad modified / exact matches (from researches competitor keywords)...
The ad ctr was optimized with stuff like keyword placeholder, best broad ad was around 1% ctr. Also tried to create specific ads to niche adgroups, eCPA was not getting really better...
So i think we tried most of known ways to optimize this campaign with keyword options / ad variations.
Best working was tablet targeting, desktop and mobile were pretty much the same...
Just used adwords search...
PPV Targeting was ebay + keyword. So if someone searches for "red shoes" on ebay we opened the pop and just showed results of red shoes...
@qkvaterpipe, building a brand was not the plan atm. just wanted to build an arbitrage mashine. Building a brand will be to budget intensive atm (employ content writers / compete with a high competitive search brand market)
@gunnar i will write you
Can you define and give examples please of what you mean,"We tried many types of campaign structures"?
How did you group the keywords in ad groups? Tightly themed or losely
Does this mean brand competitor keywords? "(from researches competitor keywords)..."
Can I assume you are getting conversions and it's just too expensive at present?
When you review your search query reports are the terms relevant for what you are targeting?
When you review your device reports which device (Desktop,tablet or mobile) is getting the best traction for conversion and ad ctr?
This will be your starting point by device, if you have nothing to go by still go with the one with the largest volume of traffic giving you more scope for growth.
In relation to PPV, did you do any url bidding?
Example, campaign urls;
ebay.de/category/baby/prams
ebay.de/category/baby/pram
ebay.de/category/baby/pra
ebay.de/category/baby/pr
And so on
This channel is a whole other universe, first focus on 1 media channel unless you have someone who is going to stick on this and you can do both Adwords & PPV. Adwords traffic quality tends to be better and you get refunds if there's fake traffic.
Some suggestions
Forget about broad match and stick to a few broad modified for gathering new keyword which can also be exact matched.
Focus on 1-2 core category products max and 1 device type for now?
Example Industry1: Baby
Specific Category1: Baby prams/ strollers/ buggies/ pushchairs
Example Industry2: Baby
Specific Category2: Baby car seats/ booster seats/
Baby stuff seems like a good start as common knowledge is people are still having babies in your target region and there's always demand for baby products and it's not seasonal.
If budget is constrained on your side, then it's important you keep focus on a core category before setting up in other categories.
Master your chosen category till you get results then it's easier to build out later on, basically like how Amazon did.
Nobody likes to hear this however, from the sounds of it, you have to work on your ads to get people over first. This in turn will increase your ctr then usually lower your cpc getting you closer to your goal.
This is something you have to keep testing, it's not 5-7 ads and done, this has to be continuous in connection with significant data to each ad.
I don't know your ads however this is all I can say from the outside.
The angle of your ad/s has to speak to your customers with either price, their concerns if they don't come to you, the positives or any social commentary being said you can use and why your customer would be mad not to go through you.
The keyword insertion for ads is cool, however fundamentally you want impact fast with a few varying styled ads that speak to your ideal user. You have to try completely different looking ads with different angles. This is likely to give you more of a chance to find better performing ads faster.
Personally I rarely used keyword ad insertion as the way I build campaigns and ad groups is with tightly themed keywords allowing for the theme of the ad group to be in the ad which tended to get better results for the group. For me this method of campaign organisation works better for reporting and managing (bids, device, keyword negatives, ad scheduling, day parting and more) without affecting other themes that maybe performing better or worse.
As one guy I heard say on a podcas, you got to make your skeletons dance about why your customers should go to your landing page whether they are bad skeletons or good so your ctrs helps decrease your cpcs.
Example,
You got a bunch of crap reviews
Yeah I got crap reviews because I'm getting a heck of alot of customers and I will get some bad ones sometimes, however it gives me the opportunity to fix problems in the market first.
Next aspect is sending users to the relevant landing page and testing which pages get you the best results from ad.
@egliptor
Campaign Structures -> tightly themed adgroups / general adgroups / specific ones / even specific ads for one keyword / passing keyword to our page / or just redirect user to a specific landingpages /....
We researched for competitors longterm keywords / scraped their titles (different type of titles from different pages) / build our own keyword list / build lists and negativ lists from analytics....
We are getting conversions, right but they are just to expensive. Its interesting that it seems not to depend on the way of advertising, in the meantime we think its our site itself as conversion price is in every way we try around 1.50 - 1.70€.
We now tried your way exactly how you descriped it for 2 weeks and suprise suprise -> conversion price is 1.76€ even after some optimization steps like analyse the search terms and made keywords out of it....
Thanks for all your help, egliptor!
I think we will now try to find another way then google...
Are you also spending any time/money on organic traffic? For a price comparison site, if you've got a genuine competitive advantage over the others, I'd expect you should be able to get a fair quantity of traffic with intelligent organic methods - viral and content marketing in particular.
Let me know what you're already doing there and I can suggest further!
Originally Posted by caurmen
Maybe to understand our plan better behind this: we are highly experienced in optimize high volume media buying campaigns. So this price comparison site was more a tool then a real product. So we bought a nice domain, contacted an agency to build this site as ofc this shouldnt just look and work like a tool and started to try CPV / Adwords.
CPV was great, scaled from day to day, we owned our own toolbars in the past so the optimization was fun and we played the game quite good - after some months we had the problem that our cpc partners gave us the feedback like:
"Okay guys cool that you are scaling and we seeing nice keywords you buying your cpv traffic and no shitty sources or fraud but your x.xxx revenue turned to 11 Sales on our partnershops - no problem guys but please optimize this in future"
Now the keywords we are buying on cpv are more longtail, big volume drop (ofc) and added intentions to all keywords like "BUY super nice smartphone for ladies" instead of "super nice smartphone for ladies". Which works fine but volume is like beeing back to the kindergarden level :P
From beginning our plan was to focus on adwords for this but as you can see in this thread we may got on this train a bit too late...
For now we had 90k costs in development and media budget for this and about 10k revenue. Close to the decision to change the status of this to -> failed with this, move on!
As this is my personal baby i dont through this away to fast but always staying realistic...
Thank you for your tips, caurmen! always open to new intelligent methods to have fun with.
Interesting - thanks for the clarification there.
Hmm - is the price comparison site all general products, or does it have a specific niche focus?
Also, do you have any particular unique advantage over other price comparison sites that you could focus on/talk about/use in elevator pitches?
These projects take a lot of money and time to get going.
Direct arbitrage is no longer possible with this model.
Returning visitors are what you are after, you need to use adwords and other high quality sources to get the initial traffic, but your site must be so good that they start to use it for their shopping research in the future. Not sure what market you operate in and how you compare to the competitors ... but you must be at least on the same level, preferable better.
With project like this, good strategy is to set an advertising budget to spend per month on traffic, and monitor the outgoing clicks to listed products. With the same budget, the outgoing clicks have to be rising from one month to the next. It takes several years to turn this thing profitable. The last similar project I was involved with, took about 3 years to start making decent profits.
It's also wise to explore other sources of traffic - we've been placing ads on classifieds sites, we've been emailing people who were looking for certain products (harvesting emails wherever possible), posting on forums dealing with specific hobbies ... there is a lot that you can do here and it's very time consuming. But in the end, it boosts the SEO a lot and that's what can kick you into profits.
Home > Paid Traffic Sources > Search / PPC (Adwords, Bing, etc.)