Determine the Quality of COD Offers (12)
Hi guys!
For the last few days, I'm preparing to start my first native+COD campaigns. In the country where I'm living (Slovenia), COD is the most popular payment method, around 75% off all online purchases are COD. I've noticed that the approval rate can vary widely.... from 25% to 70% based on the spreadsheet I got from my AMs. Anyway, I could not understand why the approval rate would be only 30% if someone already decides to leave personal data in order to get a product.
Running COD offers with CPA networks is new to me and I wanted to see how the whole pre/post-purchase looks like.
I went to AdCombo and found some offers from my country (Slovenia) and submitted a form (name + phone number) on the offers money page.
Today I got a call from a random Slovenia phone number and then I realized why the approval rate can be as low as 25%. 
Here's what happened when I got a call today.
- Lady called me without introducing herself where she is calling from
- We needed 1 minute to figure it out what's the purpose of the call
- She was talking something between Czecz and Croatian
- Shipping cost is 13 EUR for a product that costs 24 EUR 
There is no way someone would confirm an order with such bad call center service. I was working 4 years in the call center doing cold calling for a big EU company selling As Seen on TV products I know what customers on the phone don't like.
So now I'm thinking how to determine the quality of the COD offers and their call centers before even running the traffic. Here are a few ideas I came up with. Please feel free to add any input or suggestions.
- Language - Is a telephone agent speaking the same language the customer is speaking? I believe this one has the biggest impact on approval rate.
- Shipping cost - More than 50% of the product price is non-sense but I do believe it can work with some products.
- Telephone number - It looks much more legit if the number from which telephone agent is calling is from the local carrier
The challenge that I see now is how to check this stuff for different GEOs. Probably Fiverr or Upwork would do the trick.
What are your thoughts on this? Do I complicate too much? Based on the call that I got today, I think this is super important to determinate the quality before sending traffic to some random offers.
Here's what happened when I got a call today.
- Lady called me without introducing herself where she is calling from
- We needed 1 minute to figure it out what's the purpose of the call
- She was talking something between Czecz and Croatian
- Shipping cost is 13 EUR for a product that costs 24 EUR
If you're driving traffic to generate those sales, you'd better have some high converting outbound calls on the other end, otherwise it will end up breaking your campaign even if you have a strong product.
Maybe looking out for similar products in the local market and compare the experience would be a good solution in case the conversion flow can be improved in collaboration with AdCombo
The approval rate can really suffer if the call center is not good (or there are "hidden costs" like the high delivery cost).
How fast after the initial lead does the call center call the customer?
Native speaker?
How many times are they trying to get the customer on the phone before giving up?
And generally how good are their reps.
Some advertisers/operators are just better at it.
I agree with both platinum and affiliaxeoran.
Question for other affiliates
Do you check the funnel/call center quality before you start testing a new offer?
Sent from my iPhone using STM Forums mobile app
I am not running COD offers, but I have a lot of experience promoting offers in verticals which rely heavily on call center performance.
I used to test the offer flow and call center quality just as you mentioned in your post, but eventually found it counter-productive for to many reasons.
One main reason, is that it can lead me to over analyze an offer before actually testing it and seeing proof of performance (wether good or bad). On top of that it just drains extra energy and time I could be spending on testing new offers and optimizing campaigns.
I would suggest picking 2-3 offers in the same vertical and split test them under the same campaign.
This can give you actual proof of performance, and make it easier to make better decision making based on facts (rather than speculation)
Wow wakeboarder - nice testing efforts!
In some cases it can take a call center several days to call a customer, after they've filled out their information.
Many people make impulsive purchases. Give them even hours to have second thoughts and many of them would change their minds, let alone DAYS.
I'm perplexed by the fact that some of these networks aren't spending more time on training call centre staff so that they can close sales with higher success rates, or adding more people to their staff to handle the volume to decrease delays in calling customers.
And many of these call centres have hours - when the internet doesn't sleep. (It's always a good idea to ask a network for their call centre schedules for every geo - and cease traffic when it's approaching closing time / the weekend).
On a positive note: There's a huge business opportunity here for people that are willing to invest some money and time into setting up a call centre and their own ecom+COD operation.
Sorry for going off-topic - but it's frustrating to see how affiliates are spending all that money on traffic to acquire leads, only to have the call centre drop the ball. I'm not referring to any one network either.
Amy
Even though OP has great points, I got to totally agree with this. Especially since there are lot of different networks/offers in this space. Its much easier/time consuming to just split test a whole bunch of them and then pick the best ones. After that you can do the flow test if you really want to, in order to give feedback to the network.
Originally Posted by daanja
Originally Posted by vortex
And it would not even be such a big investment. If your campaign is currently working with your network, all you kind of have to do is hire someone local. Source the products and rip their landers.
Then again, the time it takes you, would be used more productively scaling in other goes. So if you have time+ resources go for it, but if you are already limited in your time, id suggest just scaling.
And it would not even be such a big investment. If your campaign is currently working with your network, all you kind of have to do is hire someone local. Source the products and rip their landers.
Then again, the time it takes you, would be used more productively scaling in other goes. So if you have time+ resources go for it, but if you are already limited in your time, id suggest just scaling.
I sent those dotpoints to my AM there and she was quite surprised and going to have a bit more look into it, we'd never done the due diligence of actually seeing what the call centre was like lol, we just asked questions and hoped the answers were right haha.
Approval rate will never reach 100%, people love shopping in the heat of moment but regret after, same as Trials and chargebacks. The approval rates also hugely swing geo to geo, Netherlands in our experience has always been very high where as Spain is much lower and inconsistent.
**for those running COD ask about guaranteed approval rates, if you have enough data to the network they can probably assure you a certain rate so you have less headaches managing campaigns on scale**
(And if you have proven track record promoting COD offers, showing stats from another network to new networks you're wanting to work with may already be enough to get you guaranteed approval rates. I've had networks offer them to me upfront, so having a good reputation on this forum helps also.)
Amy
I would just test the offers, either they convert or they don´t.
Also you would need to check the funnel several times for each offer to get a good impression of the flow, especially the call center performance.
Maybe you got a call from someone with bad mood that day, someone unmotivated so you would need to check again if that call center behavior is normal or if it was a individual case.
Then you would need to wait for the calls what also can take some time.
In all that time you could just have splittested several offers and see their REAL performance.
Also your funnel and call center testing would only base on your very own experience and doesn´t necessarily tell anything about the REAL performance of the offer.
And when we don´t only talk about COD offers but offers in general, how much time would be wasted when you check and test each and every offer funnel you run?
Dating, SOI/DOI, Sweeps, Casino, Crypto, whatever.
I never test or check the funnels, if anything I check the offers landers but most probably I just run traffic to the offers and test them.
Sure that costs money but with good planning and execution you will get that money back when you find a winner.
Originally Posted by twinaxe
But this type of investigation is still interesting to see, and more interesting to discuss. And it's really inspiring from a business point of view for anyone thinking of doing ecom.
Amy
Home > Paid Traffic Sources > Native