Home > Paid Traffic Sources > Push Traffic

---

Different Activity levels = Different quality on the back end (23)

---

I have started testing some push traffic for the first time and have noticed an interesting pattern.


The traffic is Propeller Push, I created 3 campaigns. Low Activity, Medium Activity & High Activity. All with different corresponding bids. The exact same creatives and landing pages for all 3.


We are running a CPI offer, thankfully the back end revenue data is very transparent to us.


I would think the push activity levels would affect bids, CTR, CVR, CPI, etc. But I didn't think they would effect quality on the back end.


For example, low activity has low revenue, medium activity has medium revenue and high activity has high revenue (in relation to the other 2 activity levels)


The Low Activity campaign has 14% of the revenue of the High Activity campaign.


The Medium Activity campaign has 48% of the revenue of the High Activity campaign.


This is all based on revenue per install. Overall about 400 installs so far.


Has anyone else noticed a similar pattern?

---

Very interesting thanks for sharing!

---

I don´t have such insights and usually I can imagine that the quality also reflects in the backend quality.

That being said, I also had high payout offers like CPS or Crypto CPA successful running on low activity/user freshness.

On one trafficsource that doesn´t even allow freshness or activity targeting so that I had to buy all traffic quality I even got so good traffic for a CPI offer that the advertiser gave me the highest payouts that an affiliate ever received for that offer.

So I guess it would be good to have more tests for different offers to get a clearer picture about it.

One offer/test is too less to make a final statement.

The problem is that it´s probably hard to get enough of such stats.

Trafficsources usually don´t know how the offers perform in the backend, CPA networks usually don´t know what activiy level or freshness sends what traffic.

Best would be when networks or direct advertisers who run Push traffic themselves could chime in and give their 2 cents to the thread.

---

It does make a lot of sense actually... fresh & more responsive users should definitely be more valuable. They might be new to the internet, so they consume all the content out there and subscribe to everything out there. Once they learn their tricks and workarounds, they won't fall for the offers that easy anymore. They might be new to the push concept, so they click like mad and explore all the offers... I've seen a similar situation when I was promoting app installs heavily, all the new GEOs performed like mad, especially 3rd world countries with new smartphone users, those guys would literally install everything

But I'm surprised that the difference is THAT big, I would expect maybe 10-20% increase with the highest activity and freshest users, but 14% to 100% thats a HUGE difference.

And as twinaxe mentioned, it's possible that what you are seeing is just an isolated case and it won't be the same in other GEOs or with different offers.

---

Thanks for the feedback guys. We will keep running for the moment and in a week or two I will post back the updated revenue figures.

It should be really interesting to see if the revenue figures hold!

---

I'm pretty sure the effect will be there, even with larger data sample, but I expect the difference to become lower... looking forward to your results, once some time has passed. Please post an update here once you have a larger data sample.

---

Would be great to have some more cases of different offers, maybe different verticals, flows, geos or so.

Basically different setups that make it easier to see if there is a pattern or not.

Apart from that I think Matuloo could be right with this as well

---

Here is some updated stats.


The Low Activity campaign has 17% of the revenue of the High Activity campaign.

The Medium Activity campaign has 64% of the revenue of the High Activity campaign.


This is all based on revenue per install.

---

Thanks for the update.

Is it still from the same offer or did you test different offers in the meantime?

---

So the differences got somewhat smaller, as expected, but damn... the difference is still HUGE, especially with low activity traffic.

BTW: how big is the traffic price difference between these activity levels in your case?

---

Same offer. Next month I should hopefully have a new offer & funnel to test.

---

I just bidded the suggested CPC bids.

Low 0.03

Medium 0.042

High 0.08

Once optimized the CPI was actually lowest on the High Activity campaign.

---

That's great actually... nice example of why it makes sense to pay more for the higher quality traffic.

---

I thought I would give an update on the recurring revenue of this offer. Revenue on the Medium Activity level has now overtaken the High Activity.

High Activity has 82% of the revenue of Medium
Low Activity has 18% of the revenue of Medium

I did intend to test other offers, but it hasn't worked out as yet.

---

Hm, so now medium is winning over High... the good thing is that low activity is still behind a lot, so it confirms the theory of reaching better backend results when targeting the more active users.

Did you change your bid levels significantly, or are you still following a similar structure to what you posted earlier, such as:

Thanks for updating this thread with fresh results, much appreciated!

---

Thanks for the update.

When you run medium activity do you target only medium then or medium and low?

Same goes for high activiy, is it only high activity then or high + medium + low?

---

Great insight, but is there any more worth in continuing to test low activity at this point after obtaining enough data?
Also, I don't particularly understand why you chose to divide revenue by explaining it in % of one campaign from another, it feels rather confusing I think. Just a wild idea, but for the sake of better visibility in your tests as well and in updating people here in the thread, maybe throw in a pie chart that shows each campaign and its total % of the 100% total revenue of the offer spread over all 3 campaigns?

---

Well, he´s only talking about revenue but not about profits.

So even when "Low Activity has 18% of the revenue of Medium" the low activity campaign can still be profitable so that it can be worth it to keep it running.

---

Guys, this is about the "backend revenue" not initial, so how much the advertiser makes on the backend. Higher revenue on the backend actually means better quality leads. That's what this thread is about

Based on the data provided, it's much better to focus on high&medium activity levels in order to make the advertiser happy, get better payouts or possible higher revenue share.

---

Exactly, so just to clarify this is strictly about backend revenue. The campaigns have been paused for a few months now. I am just updating the recurring revenue stats.

All activity levels were broken into their own campaigns and bids were adjusted accordingly.

In this specific situation Low Activity users would just not be viable, even at the much lower bids. The overall quality is vastly inferior to the Medium & High activity users.

---

Right, so even with the lower bids, the low activity level was performing too poor for it to make sense to run it. Thanks for the update.

---

Did you run the campaign only on Propeller or on other sources with user freshness targeting as well?

---

Interesting, I tend to agree with twinaxe, we have advertisers who have great success with fresh users only, yet some are getting a LOT of conversions from users who were collected by us over 120 days ago, the ROI is usually high as the CPC is significantly lower for these feeds.
Did it happen with only one campaign? few campaigns?

---

Home > Paid Traffic Sources > Push Traffic