Which source would you put eight hundred quid of native through?

Been on search and push for four years and I've never bought a native click in my life - which is starting to look like a gap rather than a preference.

What I've got: £800 I'm prepared to lose, an offer that's been steady on push for eleven months, landers that already convert, and a tracker I trust. What I don't have is any idea which source to walk into first. Every writeup I read is from someone spending twenty grand a month and the advice is clearly written for them.

The minimums thread on this board scared me a bit - if the real floor to learn anything is two grand then I'd rather know that now and save up than dribble £800 away learning nothing.

So: same money, same position, where are you putting it? And is the answer "don't"?

don't. eight hundred on native buys you noise and a strong opinion based on noise

Do it!! I started native on like $500 and it was the best decision I made last year

Did it make money, or was it a good decision in the sense that you enjoyed it?

second one probably. it broke even eventually

£800 is enough to learn one thing. It is not enough to learn native. Decide which one thing before you spend a penny and it stops being a waste.

What I'd do in your position, and roughly what I did in 2021:

1. Don't go to a tier one source. The minimums are real, the auction is full of people optimising against you with a hundred times your data, and you'll spend the whole budget on the learning phase of an algorithm that never finishes learning. 2) Pick a mid-tier source with a low deposit, ideally one where you can whitelist site IDs by hand rather than trusting automatic optimisation. 3) Buy on a handful of widgets, not broadly. Twenty site IDs, not two thousand.

The thing you're testing at this budget is whether your offer and angle work in a native context at all — different intent, colder traffic, different creative language than push. That's a yes/no you can get for £800. What CPA you'd eventually run at, you cannot.

One warning: on the mid tiers half your spend at the start goes to placements that are technically real and practically worthless. That's not the source cheating you, it's what unoptimised inventory looks like, and it's why the blacklist matters more than the creative for the first fortnight.

Disagree with mid tier for someone new. Is cheap traffic yes but you not know if the problem is your creative or the traffic is garbage, and you spend the money learning to read garbage.

Better take the smaller of the two big source, deposit minimum, run one campaign, one geo, three creative. Traffic is expensive but is real, so when it not convert you know is you. This is worth more than the money you save, anyway with 800 you not have time for two lessons.

That's a fair objection and I don't think either of us can settle it. Clean signal on expensive traffic against more data on dirty traffic — I've watched both work and both fail, and the deciding factor was usually how disciplined the person was about the blacklist, not the source.

Numbers so you can size the test rather than argue about it.

At native CPCs in the £0.03 to £0.12 range depending on source and geo, £800 buys you somewhere between 7,000 and 26,000 clicks. If your offer converts at 1% you're looking at 70 to 260 conversions, which is a real signal. At 0.2% you get 14 to 52 and you have learned nothing at all.

So the question that decides your whole budget is what your push CR is now and how much of it you expect to keep on colder traffic. Mine dropped by about 60% moving the same offer push to native, first attempt.

	Push	Native, first month
CR to lead	2.8%	1.1%
CPC	£0.09	£0.06
CPA	£3.20	£5.45

It got to £3.90 by month three.

Push CR is 3.1% on the offer I'd be testing. So if I take your 60% haircut I'm at 1.2ish and the £800 gets me a real number.

That's the most useful thing anyone's given me, thanks - I was about to size this by gut.

Good number. Now double it and you have a real plan.

Everyone has answered "which source" and the honest answer is that at £800 the source is close to irrelevant compared to two things you haven't mentioned.

First: the pre-lander. Push traffic will click through to a form. Native traffic almost never will, because the intent is completely different — somebody clicked a headline out of curiosity and you're asking for a phone number four seconds later. If your existing landers are what converts on push, they will underperform on native by a factor that has nothing to do with which source sold you the click, and you'll blame the source. Budget one of your creatives' worth of effort into an article-style pre-lander instead.

Second: how you'll cut. £800 across twenty site IDs is £40 each, which is not enough to judge any of them individually. You'll be tempted to blacklist on 200 clicks and no conversion, and at a 1% CR that's a coin flip. Decide the rule in advance — I use "cut at 3x the clicks it should take to see one conversion, no exceptions" — because deciding it while you're watching money leave is how people cut the winner on day two.

Source, if you're making me pick: the smaller of the two large ones, per vlad. But it's the third most important decision here, not the first.

the pre lander thing is 100% what killed my first native test and I never worked that out

Spent it. Reporting back because I always want these and nobody ever posts them.

Went with the smaller big source, one geo, one offer, four creatives, plus an article pre-lander I wrote myself over two evenings. £780 of the £800 spent over 19 days.

• 11,400 clicks at £0.068
• CR to lead 0.9%, so 103 leads
• CPA £7.57 against £3.20 on push
• Blacklisted 6 site IDs out of 34, using the 3x rule, and forced myself not to touch anything else

So it lost money and I'd call it a success. The pre-lander was the difference between 0.9% and something I suspect would have been about 0.3% — I ran 90 clicks straight to the old lander at the start and got one lead, then stopped.

One creative did 60% of the leads. It was the one I nearly didn't upload.

Next £800 goes back into the same source with three variants of that creative and the same pre-lander. Not going to pretend I've learned native. I've learned that my offer works on it, which is the one thing Marta said I could buy.

fair. you spent it on a question instead of a hope, that's the difference