Home > Mobile > Follow-along Campaigns

---

Journey to Profit with Mobile Pops & Sweeps (40)

---

Hi Everyone

I'm launching a follow along in a hope to learn faster and meet some cool people. I've also come up with a list of objectives which I'll aim to tick off over the course of the follow along.

Objectives:

• First Conversion (Update: Achieved Day 1 - 20 March)
• -50% ROI Campaign
• Break-even Campaign
• First $xx / day
• First $xxx / day

---

Traffic source: Popads
Vertical: Sweeps - Short Form Submit
Offer(s): Win an iPhone 6
Geos: Spain

Budget: $50 / day

Campaign structure:
I've started with a single campaign to test 5x landing pages and 2x offers from different networks. Objective is to see what, if anything converts, and iterate quickly.

Popads Targetting:
I've set up the following targeting on Popads. Some of the settings, such as Top 50% of Websites and Premium Spot Only are based on testing I've done in other geos where these have performed better, hence using these to hopefully get better results faster and then broaden targetting to test other areas once break-even or profitable.

• Top 50% of Websites
• Premium Spot Only
• Popunder Only
• All categories (except adult)
• Spain
• Smartphone / Tablet
• STM Blacklist

Landing pages:
To get an initial sense of the landscape I've ripped 5x landing pages from Adplexity.

• 001 - Facebook choose your prize theme
• 002 - Attention you have won theme (similar design to offer page)
• 003 - Quiz theme
• 004 - Whatsapp wheel spin theme
• 005 - Generic wheel spin theme

Offers:
The following offers were the only two Spanish Win iPhone offers I could find in the networks I've joined. I'm hoping that at least one performs.

• A - Win iPhone 6 (F5 Media) - Payout $1.20
• B - Win iPhone 6 (Adsimilis) - Payout $1.10

Actions I took:
Set up campaign, ready to launch 12pm (local time to the campaign) and run $60 with spend capped at $5/hour.

Next Steps:
Launch campaign and report results once spend is complete.

---

I actually started this test last night, so I have the results for day 1 already. Here they are...

Spend for the day: $60.08
Revenue: $9.20
Net profit/loss: ($50.88)

Detailed Results & Observations

1. Landing Page:

• -82.16% for 001 - Facebook choose your prize theme
• -100.00% for 002 - Attention you have won theme (similar design to offer page)
• -51.61% for 003 - Quiz theme
• -100.00% for 004 - Whatsapp wheel spin theme
• -89.81% for 005 - Generic wheel spin theme

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:


Observations / Next Steps:

• Landing page 003 beats all the other landing pages hands down. Even when running 003 vs 001 (second best performer) the results come out at 88% to 12%, so I think this is enough to cut everything but 003.
• My guess is that 003 won because it has a few different sub-pages loaded using JS before pushing the user through to the offer. I think the extra commitment required by the user on the landing page is helping increase conversion as the offer is a short form submit.
• Two of the landing pages had CTR > 100% - initially I was confused by this, but I now think it's a result of those two pages spamming the back button with my tracking url (http://xxxx.voluumtrk.com/click).
• Need to test new landing pages against current winner in attempt to improve ROI

2. Offer:

Both offers received 4 conversions, so not much to report here.

Here's the Voluum screenshot:


Here's a screenshot showing offer performance for the top performing landing page:


Observations / Next Steps:
Will need to continue running both offers for the time being, not enough here to pick a winner.

3. Traffic Source (Popads):

Bid: 0.00322 -> increased to 0.01033 after one hour with only $0.60 spend

Initially I set my campaign up around 7am (Spain time) ready for it to start at 12pm (Spain time). I set a bid of 0.00322 and it showed high traffic estimates (see screenshot below). However after running for the first hour, the campaign showed minimal spend and I had to up the bid significantly to get traffic. I'm guessing this is due to low demand early in the morning.

Screenshot of initial traffic estimates at 7am. Had to increase the bid significantly at 12pm to get any traffic (forgot to take another screenshot when I increased the bid)


Observations / Next Steps:
I'm happy bidding above average for the moment to get traffic as I refine my lander/offer combination. Once I've done that, I think I'll take the advice I've seen from Amy in similar follow alongs and test different bid amounts.

4. Other Observations:

I tried drilling down on a few different variables. Nothing too interesting stands out so far. I've included a screenshot below of a breakdown for the top performing landing page by offer and OS. I should also note that across all landing pages iOS and Android were even at 4 conversions each.

Screenshot of Offer/OS breakdown for top performing landing page


Summary

Overall, I'm not that happy with results thus far, however I think the poor performance from many of the landing pages was due to pushing users through to the offer too quickly, before there was enough commitment to complete the short form. If I can increase performance with a new batch of landers, the additional data may highlight profitable segments of traffic - OS/Browser/Device/etc

Next Steps

• Create a new batch of landers to test against the winner
• Continue testing both offers
• Report back tomorrow with results

Of course - being a follow along, I'm keen to get feedback and connect with other people who are on the same journey, so don't be afraid to say hi

---

Today I added 7 new landing pages to test against yesterday's winner. Here's how it went...

Spend for the day: $60.01
Revenue: $12.60
Net profit/loss: ($47.41)
ROI: -79.00%

Detailed Results & Observations

1. Landing Page:

-69.39% for 003 - Quiz theme (Yesterday's winner)
-100% for 006 - Quiz theme 2
-53.93% for 007 - Quiz theme 3
-41.74% for 008 - Quiz theme 4
-83.70% for 009 - Choose Your Prize
-100% for 010 - Quiz theme 5
-100% for 011 - Quiz theme 6
-84.29% for 012 - Quiz theme 7

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:

• Top performing lander is 008 with 4 conversions.
• Comparing lander 008 directly with each of 006, 010 & 011 (all zero conversions) it shows 008 beating each lander 98% to 2%. Therefore these three landers can be cut.
• When comparing lander 008 with each of 009 & 012, it shows 008 beating each 90% to 10%. I realise that this means 1/10 times in each comparison the other lander would win, however at this stage, I think its enough of an indication to cut both these landers.
• That leaves three remaining landers - 003, 007 & 008 which at this point there is no clear winner.

Here's a screenshot from the peak conversion calculator for the top three landers - will leave these running for the time being.


Observations / Next Steps:
Seeing -41.74% for the top performing landing page is much better than yesterday, however it seems like there's a lot of work remaining to get this campaign profitable. I'm going to focus on the top three landing pages now and create iterations of each for further split testing.

2. Offer:

Slightly more conversions yesterday (up from 8 to 11), however the split between offers is still even. Results came in 6-5 today (4-4 yesterday), so they are even thus far.

Here's the Voluum screenshot:


Here's a screenshot showing offer performance for the top performing landing page:


Observations / Next Steps:

• I ran the offers through the Peak Conversion calculator individually for each landing page. For 003 the Adsimilis offer is ahead 92% to 8%, however I'm not ready to cut an offer given how close they are across the board.
• I'm going to continue monitoring offer performance for the time being. I've also noticed another lower payout offer for Spain which I may add into the mix, however it will have to be after the next batch of testing, as I need to request cap from my AM.

3. Traffic Source (Popads):

Not much to report here as I kept the bid the same as yesterday. I'm going to split the campaign into three for the next round of testing and run low/med/high bids to see if there's any impact on overall performance.

4. Other Observations:

I've tried drilling down on a few different variables again and there's really not much green at all. This concerns me a little as most of the other campaigns I've run, I start to see some green for OS/Browser/Device/etc that shows some promise. At this point I can't cut my way to green so will need to focus on improving the funnel further.

Here's a screenshot of the three top landers broken down by offer and OS.


Summary
Overall, the first couple of days have been pretty tough, this campaign will take some work. However I'm happy to see the top performing lander at -41.74%. I think if I can push this up to around -20%, I may be able to cut may way into green.

I really wish I had some more offers to test. Hopefully I can get that third one on board shortly, but I'm not sure where else to look to find win iPhone 6 offers in Spain. Maybe my offer choice isn't that great? If anyone has thoughts on this, let me know.

Next Steps

• Run the top three landers, plus three new landers that are variations of each of these, in attempt to improve funnel.
• Continue running both offers. Research new offers that can be split test.
• Test low/med/high bid amounts.
• Report back tomorrow with results.

Feedback
I'm maintaining complete transparency with this Spanish campaign as the objective is to learn as much as possible (and hopefully that makes it more helpful to everyone else also!). If you have any feedback on my approach and results so far, let me know! I'm keen to hear it!

---

Hey! If you ripped your landers anyway you might want to show them here, or at least some stats on them since that is a highly likely "bottleneck" for your entire campaign! If a lander with -50% ROI turns out to have crappy bloated code / copy and unoptimized images, random scripts stuffed in head, unused scripts, not fully responsive, etc - fixing it could take it to closer to break-even or maybe even beyond! So you might want to have a closer look at each lander showing some promise and see if it can be improved before moving on to the next one.

---

Firstly, thanks ysekse - after reading your comment I went back through all the landers. I went through all of them quickly before launching, but upon revisiting I made sure they weren't missing anything (entry pop, exit pop - both happy & unhappy path, back button redirect, vibrate, non essential scripts at bottom of page, etc) and they're much better now. I've posted Voluum screenshots with the lander stats below, let me know if anything doesn't look right.

Here's today's update:

Today I continued running the top three landers and introduced one custom variation of each lander into the mix. I also split my campaign into three to access bid performance.

CAMPAIGN 1 - HIGH BID

Spend for the day: $19.99
Revenue: 10
Net profit/loss: ($9.99)
ROI: -49.98%

1. Landing Page:
003 - Quiz theme: 1 conversion (-67.63% ROI)
007 - Quiz theme 3: 2 conversions (-31.25% ROI)
008 - Quiz theme 4: 4 conversions (31.94% ROI)
013 - Variation of lander 003: 0 conversions (-100% ROI)
014 - Variation of lander 007: 1 conversions (-33.35% ROI)
015 - Variation of lander 008: 1 conversions (-67.67% ROI)

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:
Lander 008 is 90% of greater chance of beating all landers except for 007. Therefore this campaign indicates that both 007 & 008 should be kept running for the moment.

Observations / Next Steps:
It's good to have a lander in the green today. Results on the whole may be caused by other factors, such as day of week, so will continue to look at how landers can be improved.

2. Offer:
The results for offers today were completely unexpected, given how even they had been thus far.

Adsimilis: 8 conversions
F5: 1 conversion

Here's the Voluum screenshot:


Observations / Next Steps:
I ran the results through the Peak Conversion calculator which puts the Adsimilis offer at 99% likelihood of being best. Not sure what happened with the F5 offer today, will reach out to my AM to confirm that everything is still working.

CAMPAIGN 2 - MEDIUM BID
The medium bid campaign performed much better than the high bid campaign. Here's the results:

Spend for the day: $20.08
Revenue: 18.10
Net profit/loss: ($1.98)
ROI: -9.86%

1. Landing Pages
003 - Quiz theme: 1 conversion (-65.02% ROI)
007 - Quiz theme 3: 2 conversions (-34.63% ROI)
008 - Quiz theme 4: 5 conversions (66.15% ROI)
013 - Variation of lander 003: 2 conversions (-30.86% ROI)
014 - Variation of lander 007: 4 conversions (40.12% ROI)
015 - Variation of lander 008: 2 conversions (-31.66% ROI)

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:
The results are pretty inconclusive for this batch of landers. Landers 008 and 014 are ahead, however when comparing each lander to 008 (the top performer), the only one which could be cut is 003. Will need to think about what I do here.

2. Offer:
Again, the results for this campaign show the Adsimilis offer coming out ahead.

Adsimilis: 11 conversions
F5: 5 conversions

Here's the Voluum screenshot:


Observations / Next Steps:
The peak conversion calculator puts the Adsimilis offer at a 95% chance of being best, so it looks like I'll need to cut the F5 offer.

CAMPAIGN 3 - LOW BID
At time of writing, the low bid is still running, it's slowly getting traffic, but not as much as the other campaigns.

Spend for the day: $6.90
Revenue: 7.90
Net profit/loss: $1.00
ROI: 14.52%

1. Landing Pages:
003 - Quiz theme: 0 conversion (-100% ROI)
007 - Quiz theme 3: 0 conversions (-100% ROI)
008 - Quiz theme 4: 0 conversions (-100% ROI)
013 - Variation of lander 003: 2 conversions (129.82% ROI)
014 - Variation of lander 007: 2 conversions (85.42% ROI)
015 - Variation of lander 008: 3 conversions (178.37% ROI)

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:

• The calculator has the three landers with conversions beating the other landers by more than 90%. This is quite different from the two other campaigns, however it's worth noting here that the leading lander (015) is a variation of lander 008 which is leading in both other campaigns.
• The only lander which is called to be cut across all three campaigns is 003 - this is a good result, given that lander set the initial baseline to beat.

2. Offer
Once again, in this campaign the Adsimilis offer appears to be performing best.

Adsimilis: 5 conversions
F5: 2 conversions

Here's the Voluum screenshot:


Observations / Next Steps:
The peak conversion calculator puts Adsimilis at an 86% chance of being best. It's not quite 90%, however given the strong results across all three campaigns, this is enough to cut the F5 offer.

OVERALL OBSERVATIONS

• Running three campaigns to test different bids was a good experiment. The results indicate that a lower bid amount will perform better, however I'm now seeing tranches of green even in the high bid campaign.
• For the purposes of testing I'm going to continue with the medium bid campaign and I'll expand back to other bid amounts once the medium bid campaign is running profitably.
• I also will need to dig into the data and start focusing on the green OS/Browser/Device/etc. I'm going to run one more batch of testing before.
• Overall on this campaign I've spent $167.21 and made $57.80 for an ROI of -65%. Today's testing showed a good improvement and I think I'm now on the right track.

NEXT STEPS

• Cut the F5 offer. In its place I'll add another offer I've come across to split test.
• Cut lander 003 - allow the rest to continue running.
• Continue with medium bid campaign while collecting more data.
• Report back tomorrow with results.

---

I realized you did not mention anything about cutting placements. Have you optimize the placements yet? For example, cutting placements that have ridiculously low CTR (your landers are getting about 15%-40% CTR, consider cutting placements that have less than 3% CTR.) these might be placements filled with bot traffic, which are just sucking away your money.

all the best.


-Jon
Skype: jonsee88

---

Good point, I forgot to mention I'm cutting any placements which have 3x spend and no conversions. I hadn't thought to use CTR when cutting, though a quick look through those placements I've already cut shows most had low CTR of around 5-7%. Will keep an eye on CTR now when cutting.

---

Yup, you can also cut based on CTR, in case the CTR is too low to even have a chance to produce positive ROI. Its true you should worry more about the final ROI of each placements but sometimes the CTR is so low that it cant work no matter what you do. This is usually due to bad banner placement for example, but since you're running POPs its because of the poor quality traffic you are getting from some placements or its poorly designed bots. There is also another extreme, if the CTR is too high, it can also be a sign of a bot, so keep an eye on that too.

---

Awesome progress! Looks like profits is right around the corner!

When you do lander variations, do you change just some text/image/buttons or it is same angle just completely different design?

---

I haven't run many campaigns yet, so don't have much of a process yet. For these ones I changed the layout and text a little, but you make a good point about testing completely different designs for the angle, I'll have to try that.

---

So I did a crazy thing last night and I bought a ticket to Affiliate World Europe!! If you're going to be in Berlin, I can't wait to see you there!

Anyway, here's the results for today...

Spend for the day: $45.33
Revenue: $51.70
Net profit/loss: $6.37
ROI: 14.05%

1. Landing Pages
007 - Quiz theme 3: 9 conversions (8.37% ROI)
008 - Quiz theme 4: 9 conversions (9.33% ROI)
013 - Variation of lander 003: 7 conversions (-13.05% ROI)
014 - Variation of lander 007: 13 conversions (54.47% ROI)
015 - Variation of lander 008: 9 conversions (9.68% ROI)

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:
The calculator doesn't show the top lander as greater than 90% likelihood of winning against any of the other landers based on today's data. However, I also ran the data for these landers from the past 3 days (should be OK as they all ran at the same time). This data showed 013 as having only a 10% chance of beating the top lander, so am going to cut this one.

2. Offer:
I added a second offer from Adsimilis to today's testing.

Adsimilis #1 (Original Offer) - 27 conversions
Adsimilis #2 (New Offer) 20 conversions

Here's the Voluum screenshot:


Here's the results from the peak conversion calculator:
Despite the original offer performing strongly, the calculator puts the original offer at only an 85% likelihood of winning - I'll need to leave both running for the moment.

3. Traffic Source (Popads):
I'm still running the medium bid as this performed best in my bid testing yesterday. At this bid I only spent around $45 from my $60 budget, however I think I'm getting enough data to get to my best lander/offer combo.

Summary
It's a great feeling to see some green - and its exciting to start thinking about how I may be able to scale the campaign. However, before I get ahead of myself, my thinking is I need to find the best lander/offer combo and target an ROI of around 50%.

Next Steps

• Cut lander 013, leave other landers running to collect more data.
• Continue running both offers, with aim to get statistical significance and cut one.
• Research how campaign can be scaled once 50% ROI is achieved (just read this post from Charles Ngo which should help)
• Report back tomorrow with results

Questions

• It's nice to see some green, and it's tempting to rush into trying to scale, but I sense I need to focus on consistent days of low green first and getting to the best lander/offer combo. Is this the right approach, or should I be testing more, such as other traffic sources, already?
• I haven't cut any OS/Browser/Device yet, however I'm concerned that once I do too much traffic will disappear... what should I be doing here? Is the only option other traffic sources?
• Are you going to Affiliate World Berlin? Will you be my friend? lol (Only half joking )

---

Almost as if yesterday's green was too good to be true, ROI dropped significantly today. My guess at this point is the cause being either:

• Deciding last minute to test a scaling strategy - I set up two campaigns on popads at the exact same bid in an attempt to get more traffic. Maybe the same people were seeing an ad from each campaign, effectively increasing the frequency cap to 2 views / 24 hours. In retrospect I think I should have continued running just the one campaign and adjusted my bid as necessary.
• General Increase in bids - I've noticed that bids have been going up over the past few days, though given I'm still getting about the same CPV and my EPV is what's dropped, I don't think I can blame this. Either way, I'll dive into this further below.

Anyway, here's the results for today:

Spend for the day: $49.74
Revenue: $30.80 (28 Conversions)
Net profit/loss: ($18.94)
ROI: -38.08%

1. Landing Pages
Here's the results for today:
007 - Quiz theme 3: 5 conversions (-56.22% ROI)
008 - Quiz theme 4: 8 conversions (-28.17% ROI)
014 - Variation of lander 007: 8 conversions (-29.65% ROI)
015 - Variation of lander 008: 9 conversions (-38.00% ROI)

Here's the Voluum screenshot for today:


Here's the results from the peak conversion calculator:
For the peak conversion calculator, I didn't just include today's numbers, but the numbers from these 4 landing pages for the entire time they've been running in parallel.

Lander 007 - 8119 Trials, 16 Successes, 04% chance of being best
Lander 008 - 7988 Trials, 22 Successes, 34% chance of being best
Lander 014 - 8117 Trials, 24 Successes, 54% chance of being best
Lander 015 - 8066 Trials, 18 Successes, 09% chance of being best

Given the results, is this enough to cut lander 007 & 015?

2. Offer:
Today:
Adsimilis #1 - 7872 Visits, 12 conversions (Original Offer)
Adsimilis #2 - 7735 Visits, 16 conversions (New Offer)

Overall (for the period both offers have been running in parallel):
Adsimilis #1 - 16,285 Visits, 39 conversions (Original Offer)
Adsimilis #2 - 15,680 Visits, 36 conversions (New Offer)

The offers are on par at the moment, so will keep running both.

3. Traffic Source (Popads):

Bidding:
A couple of days ago I went through a bunch of bids and traffic estimates to determine my low/medium/high bids for testing.

bid 0.01033 - 29000 impressions -- high
bid 0.00933 - 29000 impressions
bid 0.00833 - 29000 impressions
bid 0.00733 - 29000 impressions
bid 0.00633 - 29000 impressions
bid 0.00533 - 29000 impressions
bid 0.00433 - 29000 impressions
bid 0.00403 - 29000 impressions -- medium
bid 0.00398 - 18000 impressions
bid 0.00393 - 15000 impressions
bid 0.00383 - 15000 impressions
bid 0.00333 - 15000 impressions
bid 0.00233 - 12000 impressions -- low
bid 0.00183 - 10000 impressions
bid 0.00133 - 5000 impressions

Looking at the number of impressions for the medium bid, it looks like bids are going up:

Day 3 (22/3) - 0.00403 - 29000 impressions
Day 4 (23/3) - 0.00403 - 19000 impressions
Day 5 (24/3) - 0.00403 - 16000 impressions

At the same time it looks like traffic may have decreased slightly. I used $10/pop bid to see how much traffic top spot would get:

Day 3 (22/3) - $10 - 33,000 impressions
Day 5 (24/3) - $10 - 22,000 impressions

I'm not sure what to make of all of this... I think I should be aiming for +ve ROI on a bid that gets me 10,000-20,000 impressions per day. Is there a general rule to work off here?

Running Two Campaigns:
I think running two campaigns at the same bid was a bad idea in retrospect. I can see that EPV dropped from around 0.003 yesterday to 0.002 today. I can't be certain that running two campaigns like this was the cause, however I can test that by running just the one again in today's test.

Summary
After yesterday's green, it sucked waking up to red today. However, all is good as I'm learning a bunch and hoping that tonight's test will steer things back in the right direction.

Next Steps

• Run just the single campaign tonight to see if that returns things back to profitability.
• Continue running both offers as they're performing equally well.
• Continue running lander 008 and 014 - not sure if I can cut the others, based on the data above. Will do some more research, or hopefully someone can make a suggestion?

Questions

• Based on the data above, can I cut the lowest two landers?
• How many days does it usually take to determine whether your campaign is a winner? Is this the usual process, or could I have done things better to speed up my progress?

---

i believe this is what you are comparing. Attachment 10582

to decide if you should cut 007 and 015, you can also do a heads up comparison it with the best lander
Attachment 10584
Attachment 10583

then decide if you should cut them. for me i personally think you need more data.

i also think you might be trying to split test too many things at one go. example you are testing scaling, bids, landers, offers, all at the same time. with limited budget, you would not be able to get significant data. do one thing at a time.

you can cut out the definite losers first, like those that have hardly any clicks or no conversions at all.

your campaign looks promising, hit me up on skype, we can discuss more there.


-Jon
Skype: jonsee88

---

First - thanks to Jon See (acepowermarketing) for being SUPER helpful and very generous with his time to provide feedback on this campaign. I'm so grateful for the opportunity to meet people like Jon via STM - it's a great forum!

Based on feedback I've received, I'm going to focus on the following:

1. Finding the best offer
2. Getting enough traffic, such that I have adequate data to make decisions and optimise
3. Quickly cutting poor performing placements
4. Finding the best landing page

1. Finding the Best Offer
I've only tested 3 Spain 'Win iPhone' offers from two different networks. At the moment, I don't have any similar offers for networks I'm in, so I'll need to find more offers (probably though spying) and then apply to the network so I can run the offer.

2. Getting enough traffic
I need to make sure I'm getting enough traffic to enable $50 spend per day so I can collect enough data. Previously I was having problems with traffic dropping - I'm going to attempt to address this through a combination of increasing bid and opening up to more than the top 50% of websites on Popads.

3. Quickly cutting poor performing placements
I'm now using a combination of CTR (<10% for placement) and CR (2-3x spend with 0 conversions) to access whether a placement is likely to perform and quickly blacklisting those which don't show promise.

4. Finding the best landing pages
This is still important, but it's number four on the list as I'm down to 4 landing pages and they're all performing roughly the same. I'll cut them over time as they can be separated, however I think focusing on the first three items on the list will speed up my progress the most.

Here's the update for Day 6

Spend for the day: $65.23
Revenue: $30.80 (28 Conversions)
Net profit/loss: ($34.43)
ROI: -52.78%

1. Landing Pages
Here's the results for today:
007 - Quiz theme 3: 8 conversions (-47.19% ROI)
008 - Quiz theme 4: 9 conversions (-39.66% ROI)
014 - Variation of lander 007: 5 conversions (-65.39% ROI)
015 - Variation of lander 008: 6 conversions (-59.43% ROI)

Here's the Voluum screenshot for today:


Here's the results from the peak conversion calculator:
I ran the numbers from the beginning for each of the landers - still the top performing lander doesn't beat the lowest performing by more than 90%, so not cutting any landers for the moment.


2. Offers
Adsimilis #1 - 11,162 Visits, 13 conversions (Original Offer)
Adsimilis #2 - 11,173 Visits, 15 conversions (New Offer)

The offers are performing similar again today - they're also even overall, so nothing I can cut here.

3. Traffic Source (Popads):
Only change here today was to change quality to Top 70% of Websites to increase traffic. I'm also keeping the bid around average or slightly above.

Summary
I'm not sure what's caused things to drop off to -50% again... Am feeling a bit stuck. I'm working on getting into other networks so I can split test more Spain offers, however I'm still unsure what's causing the massive ROI change. On the bright side, now that I'm getting more traffic again, that should help with finding placements that work and getting an idea of the EPV I can expect. Potentially I can curve fit bids to EPV...

Next Steps

• Continue running all landers
• Continue running both offers
• Find new offers in Spain to test
• Report back tomorrow

Questions
All offers and landing pages are pretty even. I've bounced from -ve ROI to +ve ROI and back to -ve ROI again. Feeling a bit stuck... what should I be focusing on from here?

---

Hello

the fact you dropped to -50% could be just a bad day, when this happens to me, I try not to touch anything and the performance usually improves the next day, so Id suggest to NOT jump into some quick optimizations etc... cause you will screw up the campaign.

Since you were already moving around the breakeven point if I understood it correctly, you need to improve something to get into profit - a bit here and a bit there, it will add up. Not sure how much conversions you already sent the network but try to ask for a bump after some volume. You could also pick just the best LP and try to make some variations of it. The last step would be to play with bids, try to decrease it a bit to save some $, but dont decrease too much so you dont loose the volume. Looking for a better offer is also always a good idea.

Let us know if the conversion improved again and if it was a weird bad day as I suggested.

---

The drop in performance continued today, here's the numbers...

Spend for the day: $70.11
Revenue: $31.90 (29 Conversions)
Net profit/loss: ($38.21)
ROI: -54.50%

1. Landing Pages
Here's the results for today:
007 - Quiz theme 3: 9 conversions (-43.16% ROI)
008 - Quiz theme 4: 6 conversions (-62.33% ROI)
014 - Variation of lander 007: 4 conversions (-74.65% ROI)
015 - Variation of lander 008: 10 conversions (-38.27% ROI)

Here's the Voluum screenshot for today:


I ran the results through the Peak Conversion calculator since the beginning, they're still too close to cut another lander, however it's looking likely that 015 may be cut tomorrow.

2. Offers
Adsimilis #1 - 12,826 Visits, 14 conversions (Original Offer)
Adsimilis #2 - 12,872 Visits, 15 conversions (New Offer)

Again, the offers are neck and neck, so nothing to cut.

3. Traffic Source (Popads):
No changes here today, I'm continuing with the original strategy of bidding average or slightly above and cutting placements which aren't performing.

Summary
ROI is down significantly still after hitting -10% on day 3 and +15% on day 4. It seems to be running more consistently now between -40% and -50%. I'm going to continue running the campaign and see how things play out over the next few days.

Next Steps

• Continue running all landers and offers until statistical significance is reached and they can be cut.
• Find new offers in Spain to test (this requires access to new networks, so will have some lead time).
• Report back tomorrow

---

Here's my stats for the 7 days thus far..

Day 1: -84.69%
Day 2: -79.00%
Day 3: High Bid -49.98%, Medium Bid -9.86%, Low Bid 14.52% (continued with medium bid - which is average bid to slightly above)
Day 4: 14.05%
Day 5: -38.08%
Day 6: -52.78%
Day 7: -54.50%

It's three days of low ROI after the campaign showed promise on days 3 & 4. Funnily enough, I just check the stats for Day 8 which is a couple of hours in and they're green so it will be interesting to see how they look at the end of the day. I'm not making any major changes or cutting anything other than poor performing places (2-3x spend with no conversions or 100 views with no clicks).

Is it normal for a campaign to have such large swings? I was surprised it dropped from +14% to -50% within a couple of days.

I'll try asking for a pay bump and a quality check (I've seen this recommended in other follow alongs). Is there a standard number of conversions or revenue amount at which you can ask for a pay bump? What's best practice?

I'll try more landing page variations if the ROI bounces back a little.. if it doesn't I'm thinking I may need to test new offers or completely different angles. Or maybe something else that I'm missing?

After a couple of hours of day 8 it's in the green - which is a much better start than the past three days, all of which were deep in the red at the beginning of the day. It was three weird days.. but lets wait and see what the stats look like at end of day 8.

---

Hm, now it looks like the positive ROI was actually an exception, not the last performance drop

Anyways, the swings are pretty normal with low volume, the more traffic you run, the more stable it usually becomes.

As for the bump, I usually start to ask for these when I send stable volume for a few days. The amount of daily leads depends on how big the payout is. I try to send about $100 per day or more for a few days, then ask for a bump. If there is a daily cap for the offer, I try to hit it a few days in a row and then ask for quality check and to get the cap lifted plus payout bumped. You need to show the aff network that you ask for stable volume, day by day, then it makes more sense for them to check the quality at all.

It might be a good idea to test new angles and offers, there is always the possibility that what you are using currently simply wont work.

There is also one more thing you can take a look at - dayparting. Since you mentioned day 8 is in profit so far, maybe there are profitable hours every day and you can focus on those?

---

Thanks Brad for inviting me to your follow-along! However it looks to me like you know what you're doing, and you've gotten very good advice from the community.

There are many possible reasons as to why conversion rates can drop all of a sudden. I've listed some of them in the FAQ thread under "My campaign was doing OK before conversion rates have plummeted suddenly. Why?": http://stmforum.com/forum/showthread...-AM-FAQ-Thread!

You may have just gotten outbid. If you compare the placements you were getting traffic from before and after the CR drop, you'll see whether or not this is the case. You could try increasing your bid to see if things will improve.

As long as you keep testing and improving your funnel, you'll be making progress. You'll always get fluctuations in CR, but the range of the fluctuations will shift to green as your funnel improves. So I'd focus on testing and not get too caught up on the ROI for now.

Also - try to sign up to more networks that have ES iphone offers and test more offers. Testing offers has the potential to improve your ROI by a lot. Since you've already gone through several rounds of lander testing, doing a round of offer testing may make for a better next step.

You'll probably hit green again soon - looking forward to it!



Amy

(BTW Congrats on deciding to go to AWE! Will see you there!)

---

So I continued to run the this for another couple of days. It's hovering around -30% for both days and overall.

Day 8
Spend for the day: $46.33
Revenue: $33.00 (30 Conversions)
Net profit/loss: ($13.56)
ROI: -29.12%

Day 9
Spend for the day: $46.47
Revenue: $30.80 (28 Conversions)
Net profit/loss: ($15.67)
ROI: -33.72%

Overall Campaign Stats
Overall Spend: $305.09
Revenue: $216.20 (196 Conversions)
Net profit/loss: ($88.89)
ROI: -29.14%

I've placed the campaign on pause whilst I figure out what to do next. I have a bunch of data, so I'm going to drill down and analyse everything I have thus far.. I'm also hunting for new offers to split test (I think I should have one more in the next day or two).

Also, I noticed that the iOS/smartphone combination was performing fairly well, however when targeting these in popads I'm only getting 3,000 daily impressions. I think this is because I've excluded adult traffic (offer doesn't allow adult). What are the best sources for non-adult pop traffic? I'm thinking to try adcash, propeller or popcash...

---

Unfortunately, all hours are in red.. I think day 8 just got off to a lucky start.. variance had me excited.

I'll try a round of offer testing (once I have access to more offers) and then try some more angles..

---

-30% isnt that bad anymore, that could be optimizeable into profit.

Its a good idea to analyze the data you have so far, look for any possible pockets of profit - OS, browser, carrier ... whatever that you can use for targeting in the particular traffic source. As for more sources, you pretty much named your best options with good volume : adcash, propeller and popcash + popads that you are using now. You might also want to look for offers that are not so sensible to adult traffic and get your hands dirty a bit IOS will always be lower when it comes to volume, even in adult.

Keep us posted.

---

Hey, If I understood you correctly, you are running on PopAds. How do you deal with drastic bid changes? Do you just set one constant bid, set the budget and let it run? Or do you adjust the bid mid test?

I find it difficult to find a balance there. At 6PM on a 0.0007 bid I would get about 6k pops in an hour, but at 2AM about 25k. I would spend all my budget on the times that no one else is bidding.

How do you deal with that?

---

Its not that "nobody is bidding" but rather the peak hours in GEO you are targeting These are different based on the timezones and you should try to keep the same bid and take the extra traffic in order to have balanced data. If you decrease bid during peak hours, the traffic quality will change.

---

Yeah, I get the same thing. Runs overnight for me though, as different timezone, so I've been setting the bid such that I get enough traffic by morning. Some hours I'm spending around $0.50 and others it's $10.

---

I started a round of offer testing and noticed that certain offers are performing better for each OS/Device combination. After reaching statistical significance, I've cut some offers for each combo and have all traffic running through a Flow in Voluum.

Here's the results for the past couple of days (no Day 10 as campaign was paused).

Day 11
Spend for the day: $48.19
Revenue: $45.50 (41 Conversions)
Net profit/loss: ($2.69)
ROI: -5.58%

Day 12
Spend for the day: $44.06
Revenue: $45.20 (42 Conversions)
Net profit/loss: ($1.14)
ROI: 2.59%

I can see carriers and browsers that I could potentially use to optimise the campaign further, however I'm not sure of the feasibility of this on Popads as there just isn't enough non-adult traffic for Spain.


Day 12 - AdCash Experiment
I've started some testing on AdCash - was actually pretty bad.. I'm not sure why it performed so poorly, but I did notice CTR's were down significantly compared with Popads (40% vs 10%) and I also cut a lot of placements with 0% CTR. I'm guessing things will pick up here as I continue to cut bad placements. On the bright side, there seems to be much more traffic on AdCash.

Spend for the day: $68.23
Revenue: $10.20 (9 Conversions)
Net profit/loss: ($58.03)
ROI: -85.05%

I'll see what I can do over the next couple of days on AdCash. I think I need to get things working there for the campaign to be profitable.

---

Thanks for this comment matuloo - I was almost ready to move on from this campaign, but still might be possible to get it working.

---

Ok, it's been a while since I posted anything here... I've been testing on and off, but have been lacking a proper strategy to find good offers and optimise them.

So I have developed the following systems:

• Procedure for testing offers and optimising those with promise into the green
• Campaign Launch Checklist
• Tracking spreadsheets to keep track of active campaigns and offer testing

I'm treating this as a 12 week challenge with the aim to be profitable and running $200 per day by AWE (I chose this as it's feasible, but will take a lot of testing and hard work - I feel that aiming for $500-1,000 per day is setting myself up for failure and aiming lower means I will not be pushing hard enough).

Here's my spreadsheet with my plan + motivation photos (faces blurred out because I don't think my family wants their faces posted in random forums lol):



Here's how I'm going to track day by day:



I just completed a 16 week gym program, and the one thing I learnt from that is the biggest difference in results comes from DOING what is in your plan and not missing a workout - or in this case not missing a planned campaign launch or offer test.

At 10 offers tested per week, that's 120 offers by the time AWE comes around and if I don't have something working by then, I'm really doing something wrong!

Budget for this is around $10,000 and I'll be working on this full time.

I'm not going to post every day, but I'll keep the thread updated once or twice per week

PS - here's my overall stats thus far from my first (almost) 2 months in affiliate marketing. Lots of red, but I know if I can execute against my 12 week plan, then I'll find success.

---

Looking good! all the best.

-Jon See
Skype ID: jonsee88

---

So, I've been thinking a little about what I've learned thus far and what I can improve upon...

1. Test more offers
Everywhere I keep reading that the key is to test more offers as they have the greatest impact on ROI. However, I've had a tendency thus far to get caught up in trying a single offer and trying to make it work. Somewhere on the forum I read offers > traffic source > placements > landers so I'll focus on attempting to optimise in that order. (I have notes from a conversation with Jon See about the importance of testing offers - was silly of me to not start focusing on this earlier). I have a list of 10 offers for one geo I'll be testing, ready to go next week.

2. Spend Money Wisely
For each offer I'm going to spend the greater of 5x payout or $10 to test it. I think I should be able to get a rough idea of the prospects of an offer with a small about of spend. Whilst I've only had a single green camp (offer was paused), the one thing I noticed is that it performed off the bat without much work. My theory is that a good offer will stand out in a small test and can then be optimised into green. Previously I think I've had a tendency to spend too much where there's not much promise.

3. Don't be lazy with landing pages
Ripping landers from spy tools + manual spying is good to get an idea of what works, and get baseline statistics. However, it's key to improve upon existing landers - try different themes, angles, etc even if the style (questions, spinner, etc) is the same. Optimisation checklist should help with this.

4. MINDSET
This is really freakin' important! One of the hardest things is not having met anyone who's making a living from promoting CPA offers - obviously people are and there's a whole industry built around it, but it's still a mental leap to believe that this can work (and seeing red doesn't help haha). Attending AWE will help, however I need to make good progress between now and then, so whenever I find myself doubting whether this will work or getting stressed, I'll take a break, chill out for a bit and come back to it. Shifting to more of a focus on daily actions and tracking/reporting on those will help.. even though this was always the plan, I didn't have the right framework and ended up spending too much time refreshing Voluum stats! Now each test will have a start/end time and I'll only check stats while a test is running if required as part of the test case EG check at halfway point and cut landers.

Hopefully this helps others who are starting out. If you're reading this and feel like you're struggling, I can't recommend highly enough getting your thoughts written down.

---

Thanks for the thread Brad, your persistence is truly inspiring for a beginner like me. I have learned a lot of the posts in this thread. Keep it up, and I am certain that you'll hit your targets before Berlin!

---

Wow, this plan looks awesome! I'll be following this one closely.

Nutella as motivation ... doesn't get much better than that.

---

Thanks everyone for the support - pressure is on

I've gone through my finances this morning and locked in a budget to $13,000 AUD which is $10,000 USD.. all figures from here are in USD.

Regarding monthly spend, I've been doing some calcs. Assuming worst case is -50% ROI and I can afford to lose $10,000 total, I could spend as follows:

Month 1: $10,000
Month 2: $5,000
Month 3: $2,500

And if the worst case -50% assumption holds, then I'll come in under my threshold.

However if things go worse I won't make it through the 3 months and I'm not sure I have the experience yet to run my first month at $10,000. So instead, I'm going to take a slightly more cautious approach, but still semi aggressive.

Month 1: $4,500
Month 2: $4,500
Month 3: $5,500

I think I can better manage this budget and it gives me a higher risk tolerance. Of course if things go well (which I hope they do! - though hope is not a strategy, I have a better strategy than hope haha), I can always increase my budget.

So in conclusion, over the first four weeks I'll be aiming to perform $4,500 (or $150 per day) worth of testing.

---

I have been thinking through the exact same thing.

What if I lose this much or that much?
What if I have this or that ROI, how long can I go before I run my budget completely dry??!

$10k is a good budget so fear not! Worst case you just have to do a plain old job for a while and then try again! I think it's super unlikely to blow through $10k with pops and sweeps and not learn enough to hit at least one profitable campaign.

---

I'm a couple of day's in - here's a quick update.

I've tested 5 offers thus far, each with two landers. Here's the results:

Offer 1: -74.81%
Offer 2: -83.42%
Offer 3: -64.48%
Offer 4: -100%
Offer 5: -88.24%

I'm going to continue with offer 3, as one of the landers is performing at >-50% ROI. Other offers I'm leaving for now.

I've got more offer tests lined up and another is currently in progress and at -20% thus far --> I'm guessing this is the type of offer I'm looking for through this process.

Should have more for the next update.

Full steam ahead!

---

Your goal is fantastic Brad! I'll be rooting for you as I'm sure everybody else is too!

Testing massively and only optimizing the low-hanging fruit is the way to go! However - those 5 offers you mentioned in your last post for example - are they in the same vertical? Same geo?

It may save you time, money and effort if you'd focus on testing vertical+geo combos instead of individual offers. What I mean is, instead of testing an iphone 6 offer in geoA, antivirus in geoB, and wifi app in geoC, you could instead pick a particular vertical (that you know has lots of hot offers at the moment), then test several geos at a time.

So for each geo, you can ask your AMs for recommended offers, then rip a bunch of landers for an initial test. Do this for several geos at a time and optimize the one(s) that have potential (like what you've been doing).

Testing several offers at a time will cast a wider net for you than testing individual offers. And if you're testing several proven offers using "proven" landers (i.e. the ones being used the most as seen from spy tools), you can get a good indication of how easy it is (or not!) to run the vertical+geo at the moment.

Also, by using this approach you'd be split-testing offers from the start. Compare this to testing individual offers - you may test a UK gaming offer today and find it doesn't convert. Then a week from now your AM may recommend another UK gaming offer so you test that again and it doesn't convert too well either. Now you have no idea whether it's the vertical+geo just being more difficult to make green, or if it's because you happened to have tested 2 dud offers. Whereas, if you ask all your AMs for the best offers for that vertical+geo, you test them once and if numbers don't look good, you know to write it off for the moment.

For the initial testing, I'd suggest you use the traffic source you've had the most success with, and test 3 bids from the start (low-mid-high). Run 10-20x payout to each bid to get a gauge of how promising the offers are, and also to see which bid will give you the best ROI, and then just use that camp to do the rest of your optimization (pausing the other two) to save some money.



Amy

---

Yep - All from the same vertical and country. However I started with tier 1 again, so payouts of between $1 and $2.. I think this was a bad idea, so this week I'm going to mass test offers in tier 2/3 geos. Need to stop making things hard for myself.

I 100% agree. The more I test, it seems that finding a good offer and mastering the traffic source is key. I've tested a bunch of offers that show promise, but I'm struggling to get from the -20% range into green. And that's when I'm -20% without too much optimisation... I tend to find that once I start spending more money and cutting placements, I get less conversions. I think I'll try to post a case study around this so I can get more specific feedback.

This is definitely the plan, I think I need to speed up my execution though. I tested 8 offers in one vertical/geo last week, however it sounds like to do this I should be looking to do that across 5-6 geos in a week.

This is awesome advice which has strongly influenced this weeks plans. Can't thank you enough

Thus far I've been using Popads as my go to traffic source. I wonder whether that's a good decision though... how do you decide which is your go-to traffic source? I started with Popads as I didn't know where to start and other beginners seemed to be using it. I've also mostly stuck with it as a traffic source as I wanted to master a single source before moving to new ones. Should I be testing others by now?

---

Here's my first weekly update of the 12 week challenge.. Quick reminder, the aim of the challenge is to hit $200 / day in profit by the end of 12 weeks.

Week One Results
I'm tracking leading metrics - ie, if I do the below for 12 weeks, I'll achieve my goal or be really damn close!

• Test 5-10 Offers - ACHIEVED
• Record 6 hours productive time per day (Mon-Sat) - ACHIEVED
• Spend $150 per day - NOT ACHIEVED (came in around $60 per day - need to test more offers!)

On the whole it was a pretty good week. I stuck to one geo/vertical and tested a bunch of offers/landers and I found one combination that looked promising, but it didn't work out - I'm going to do a deep dive on that one below.

Deep Dive - Tier 1 Offer/Landing Testing.

Summary: I thought I found an offer/lander combo with promise, but failed to optimise. What would have you done in this spot?



I ran the above test over a 48 hour period to test 4 different offers (using two landers for each offer). At the end of the test, all offers were pretty even, however there was one lander/offer combo at -65%, so I figured I'd give it a shot. The combo only had 6x offer payout spend so it was time to keep spending/testing.



Over the next 24 hours I ran 5 landers (the best performer from above plus 4 new landers), running 6x offer payout to each. One lander that stood out massively, running at almost breakeven. This is before I have looked at cutting any placements/targets.



Over the next 24 hours I tested lander variations (same style, but different theme) to see if I could get green through lander optimisation. None of the variations beat the original, however the original continued to perform well. Over 2 days and 12x payout spend, the top lander was only a touch behind break even.



Over the next 24 hours I ran just the best lander, with the plan to get more data and begin cutting placements and optimising into the green. However, the ROI dropped massively! Not sure what caused this, but I cut placements using Amy's spreadsheet and continued to run.



The next 24 hours, the ROI dropped off further and I wasn't sure what was causing the drop. I figured that cutting placements/targets wouldn't work from -60% ROI so I paused the campaign.

Overall, I thought I'd found a winning offer/lander combination that could be taken into the green but it didn't work out.

Questions:

• Did I need to spend more finding the right offer/lander combination? How much time/spend does this usually take? I decided to optimise further off 12x spend over 48 hours, was this a good decision?
• Did I up the spend too quickly? I went from $10/day where it was breaking even to $30, then $60 in an attempt to get data more quickly so I could optimise into the green. Could have this caused the drop off? And if so, how do you optimise around this?
• Did I give up too quickly? Is the ROI drop normal at such an early stage in the optimisation process? And if so, how would have you continued with the campaign to take it green?
• Is there anything else I'm missing that I should have tried?

---

That could mean that you haven't optimized your funnel enough before cutting placements. When you say "-20% without too much optimisation", how many landers and offers did you actually test before reaching that -20%?



Haha popads is my go-to source as well! :P And also Zeropark. But yeah - popads is getting more competitive, no doubt about that. But the traffic quality is there - and that's the main thing you need when testing and optimizing a campaign.



First of all I want to say - good testing effort! There will always be ROI fluctuations - and this is exactly why we always want to be comparing more than one offers/landers - because (in the ideal situation at least) the best-performing lander will still emerge the winner in spite of ROI fluctuations.

Regarding how much to spend on each lander+offer combo - I don't have any hard-and-fast rules.

I suppose you could enter each lander-offer combination's stats into my calculator spreadsheet and evaluate them just as you would placement stats, but I haven't done it this way. One reason is that by using the calculator, we'd be considering each combo separately without looking at the trends. Example: Say I was testing the 5 most-popular landers I found through spying, and OfferX didn't convert for any of them, but some of the landers have converted for the other offers, then I'd have reason to suspect OfferX to be a dud. So I wouldn't want to just consider each lander+offer separately.

In your case though, I definitely agree with you that Offer4+Lander3 has a lot of potential! The ROI drop-off could be due to day-to-day fluctuations, or maybe you got outbid by competition. Either way, I would definitely suggest doing more testing. But before I suggest a testing strategy, I want to clarify a few things about split-testing and statistical significance:

1)You cannot conclude that lander 1 was the "better lander" because statistical significance hasn't been reached. Please see this thread for a step-by-step. Or did you make that conclusion based on lander stats from ALL 4 offers combined? If so then please disregard.

2)Landers 3 and 3a actually hadn't finished the race yet - "probability of being best" is at 84/16% and hadn't reached the recommended 90%. But if you want to make the assumption that lander 3 is best, I won't object - will let it slide just this once because there are other things you could be testing to increase ROI by greater amounts.

3)The split-testing between Offers 1 to 4 had not run to completion yet, which means if you run more traffic to them, you may find that Offer4 isn't the best offer.



And here's what I would suggest that you do for this camp:


1)All we know for certain based on your stats, is that Lander3 is the best-converting for Offer4 out of Landers1/3/4/5/6. We wouldn't know about Lander2 because you cut that out prematurely. Please check, for that stats shown in your first screenshot, the overall stats for Landers 1 and 2 (totals across all offers). Check the split-test calculator for stat sig. If Lander1 is indeed the winner, then forget about Lander2 from now on, and just run Lander3. Otherwise, run both Landers 2 and 3.

Are your 4 offers really similar? If so, based on the fact that lander3 is the best out of landers 1/3/4/5/6, we could assume the same for offers 1-3 as well.

So: Run Offers 1-4, with lander 3 (and maybe also lander 2 depending on what I said in bold above). Cut stuff as things reach stat sig, until you're left with a winning lander and a winning offer.

Sorry if it sounds complicated. Ideally you'd want to re-test all offers and landers again and cut as stat sig is reached, until you have a winning lander and a winning offer. But I'm choosing a shortcut to help save test budget without risking increasing the margin of error by too much.

2)Clone the camp and test 3 staggered bids to see which one yields the highest ROI - run 10-20x payout to each camp should be plenty. Pause the lower-ROI camps and use the remaining to conduct tests below.

3)Do another round of offer-testing with new offers (don't need to just be AM-recommended offers as long as they're very similar to your winning offer), using best lander.

4)Do another around of lander-testing using best offer. Try to test different lander styles, as well as variations of best lander.

5)There should be a lot of data collected by this point - so start cutting the bad converters! Cut placements using spreadsheet. See if there are especially bad-converting OSs/carriers/browsers etc. that you can cut. Implementing day-parting would be another good option. Just don't do all these cuts at the same time - because e.g. when you cut placements, the ROI for your OSs will increase.

6)At any point during the above, as soon as you reach a good ROI - say 30-50%+ - start scaling to other traffic sources!


All the other observations / thoughts /plan you've noted are great! And yes testing tier 2/3 geos you can find the low-hanging fruit.



Amy

---

This is absolute gold and has been so helpful - thanks Vortex!

---

Home > Mobile > Follow-along Campaigns