Hitting a wall with Meta Ads for B2B leads - agency or DIY?

datanerd

i've been running Meta ads for my B2B lead gen campaign for about four months and honestly, I'm starting to lose my mind.

We're a tiny team: two co-founders (partner still working full-time), and I'm the only one running the business day-to-day. Goal is to scale this to around $1M ARR eventually.

Worked with an agency for two months end of last year. Disappointing is an understatement. They weren't proactive-I kept having to tell them what to test or change. Design was decent following instructions, but zero strategic creative direction coming from them.

eventually i joined a growth programme and the advice was basically: "Run ASC broad and feed creatives."

So now i run:
• 1 ASC campaign with broad targeting
• heavy creative testing ($100-150/day)
• usually 10-15 creatives at a time

At the start of the year performance was decent. Conversion rate around 3% on some winning creatives.

But May has been brutal. Some creatives get insane CTRs (20-30%+) but conversion rate collapses below 1%. audience age suddenly skewed 65+ on many variations of previous winners.

i can't tell whether the problem is:
• creative angles
• offer positioning
• ASC/broad structure itself
• Meta traffic quality
• account history/data issues
• landing page mismatch
• or just creative fatigue

Every day I'm analysing old winners, iterating, testing new hooks, founder UGC, testimonials, even AI-generated clones. studying metrics, trying to reverse engineer what changed.

i wrote a quick script to pull age breakdowns from the API:

import requests
# simplified - get ad insights with age and conversion breakdown
# compare CTR vs conversion rate by age bucket
# found 65+ bucket has 4x higher CTR but 0.8% conversion

It strongly suggests audience skew or platform algorithm issues, but I'm not sure.

Question for people further along: at our stage, is it worth hiring another agency? Or is this one of those phases where founders should keep learning ourselves, get an one-time audit/creative strategist instead, simplify testing, and just keep iterating internally?

What actually helped you break through this messy early phase? Right now Meta feels incredibly inconsistent for B2B leads.

cyberpunk

Agencies usually offer a free audit. Some do paid ones, which can be marginally more helpful. Either way, they'll point you in the right direction - and you keep the findings even if you don't hire them. It's part of their sales process, but the insights are yours.

You're right to keep testing creatives. If those ads bring traffic in but conversions stall, my first instinct is landing page friction or offer friction. Assuming tracking can be ruled out. Competitors might be better priced or offer better value. Sometimes your audience is just more sensitive to economic conditions. External factors can tank performance even when nothing's wrong in the account.

Hard to say more without seeing the account, but that's where I'd dig.

One thing that's helped me: feeding all the data into Claude or another AI tool for pattern recognition. Staring at spreadsheets all day misses connections. An AI can pull insights across different sources and show the full picture.

maya-b2b

tbh, you're already doing what most agencies would do anyway

conversionninja

You're honestly closer than you think. The fact you're running tests with that kind of discipline and peering into the data like a detective? That's the bit most founders skip. They chase shiny creatives or cheap wins, never stopping to actually listen to what the numbers whisper. The brands that get it right are the ones obsessed with the details-the ones who treat their ads like a conversation with their customer, not a shouting match. You've got that instinct. Trust it.

24fire

I remember reaching out a few months ago when this thread first popped up. Didn't get a reply-completely understand, running ads for a skincare brand while questioning agency help is a headache most of us know.

But here's the thing: hiring an agency when you're already tearing your hair out often just adds another layer of frustration unless you've got the right brief. The market's moved. Meta's algorithm isn't the same beast it was two years ago. Agencies promising ROAS of 5x? Probably running budget through a funnel that only works until the next iOS update.

if you want my honest take: do a three-month trial with a specialist-someone who's worked specifically on beauty or skincare DTC. not a generalist agency. test their reporting, their attribution method, their willingness to explain why something failed. If they can't articulate that in plain English, run.

And yes, i just checked back-sent you a message in July with a couple of growth levers that worked for a similar brand. might be worth a look if you're still weighing options

trendspotter

I've been through this exact nightmare with a few ecommerce brands, so let me paint you a picture of what actually works when you're drowning in ad costs and no consistency.

First, sit down and map out every single person who could genuinely fall in love with your skincare-not just women 25-45, but the acne-prone guy, the eco-conscious mum, the luxury seeker. Then rank every benefit of your product by which ones have already scored in past ads. Write down every hook that's ever worked, no matter how small the win.

Now the fun part: storyboard a creative for every combination of audience, benefit, and hook. I mean really visualise each one-gorgeous close-ups, before-and-after shots, texture shots. That's where the magic lives.

Then you force-feed Meta these ads using ABO, one ad per ad set, and you give each one a daily budget equal to three times your target customer acquisition cost. After two days of zero conversions or a full week of missing CAC, kill it without mercy. The winners? Double their budget until they stop hitting the target, then adjust.

This isn't how I'd run a huge account, but if you've lost the formula, forcing yourself into a brutal creative testing cycle is the only way Meta will actually learn what works. If one ASC campaign was humming, you wouldn't be here tearing your hair out.

And yeah, fix your site conversion, your social proof, your CAPI-all that matters. But nothing moves the needle faster than making ads that feel like a story, not a sales pitch. Your brand deserves that.

adcraft

Your CTRs being this high is actually a warning sign, not a success metric. Meta loves shoving traffic onto the Audience Network because click-through rates skyrocket there-but the quality is garbage. Check your placement breakdown. Bet you'll find a huge chunk of traffic from Audience Network. Kill that placement and things should normalise.

i run an agency and we work with a UK skincare brand. Happy to take a quick look if you want, but honestly, at your current spend level you're not ready for an agency. Fix the placements first. that'll probably sort the issue

chatbox

Having worked with plenty of ecom stores that felt exactly this pain, I'll cut straight to it: does Meta's pixel actually know your true revenue per sale, and is that number correct in your Events Manager? sounds basic, but I can't count how many times I've audited a skincare brand and found the purchase event either firing on the wrong page, using static values instead of dynamic ones, or leaking data through a misconfigured GTM tag. get that right before you even think about agency fees or scaling budgets. If the signal going back to Meta is garbage, every optimisation decision built on top of it is just expensive guesswork.

weekendhustle

adding to what others have said, i reckon it really depends on what you're hoping to get out of it - are you after social media presence, or straight-up ad management? Both are valid, just different skillsets. if brand awareness and authority are part of the mix, I'd be happy to chat about some digital PR tactics that could complement your ads. 😊 Hope that gives you a bit to think about!

sophiasaas

I've been in that exact spot - high CTR, abysmal CVR, and the urge to throw money at an agency. Before you do, take a hard look at the message-to-landing page alignment. That's usually the culprit. Shrink your test size too, running too many variables at once just creates noise. for me, spinning up fresh statics and UGC angles with something like Krev AI cut the churn in half. You need to isolate whether it's the hook, the offer, or the product page that's sinking the conversion rate before you scale anything

chatbox

Let me tell you, seeing skincare ecommerce brands run Meta Ads without a proper tracking foundation always gives me the same headache. Advertisement is not a money faucet - it's a measurement and iteration game. From what you've described, there's no real strategy here, just "run ads and hope." That's not a strategy. I'm not saying you haven't done other work, but if the core plan is "spend £XX on Meta and see what sticks", you're setting yourself up for disappointment.

First thing I'd ask: what's your actual differentiator? Is it product formulation, pricing undercut, or unique promotion angle? If you can't nail that down, no ad platform will fix it.

Second, that budget is too lean to break into any competitive skincare market. I'd strongly recommend narrowing your targeting to a smaller geographic area. Pick one city or region, build frequency there, and actually get enough impressions per user to drive conversions. Without frequency, you're just burning cash on cold traffic that never remembers your brand.

Third - and this is where I always start - make sure your Google Tag Manager container is actually firing the Meta CAPI events properly. I've seen so many PMs waste budget because their purchase event fires on page load instead of on confirmation. Run a quick debug in GTM, check the pixel loads in the network tab, and validate that your server-side tracking is sending consistent deduped data back. If any of that is broken, your ads manager is flying blind.

Agencies can help with creative and scaling, but if the analytics layer is shaky, you'll just pay them to guess. Fix your measurement first, shrink the geo, and prove the unit economics on a £200/day budget before thinking about an agency.

emailwiz

Oh, I've been down that exact road. Spreading £150 a day across ten to fifteen creatives is like trying to water a whole garden with a teaspoon - nothing gets enough to grow. You're not going anywhere because you're fragmenting the budget into tiny, ineffective chunks.

Manual campaigns still work, honestly. Advantage+ sounds flashy but it's not a magic bullet. Some of my clients swear by it, others can't make it perform at all. You really need to give us some more info on your site and landing page though - that could be eating your spend without anyone knowing.

Best thing you can do? Pay for a proper one-time audit. Those free ones are just sales pitches dressed up as help. They'll exaggerate problems to get you on a retainer. A paid audit will actually dig into what's broken.

growthgeek

Hiring another full-service agency right now would be the wrong move. Your problem isn't that "no one knows Meta" - it's that you've got too many variables changing at once while Meta's delivery engine keeps widening and sending you crap traffic from different pools every other day.

That high CTR is the dead giveaway. Your creative is doing its job - getting the click - but either the click quality is falling apart or the landing page experience is bleeding conversions. Either way, adding more agency bloat just piles on more noise, more random split tests, and more reporting fluff.

You'd get far more value from stripping it back: one tight funnel, one campaign structure, proper attribution windows (not Meta's default 7-day click bullsh*t), and a solid operator who can audit your setup without selling you a retainer. Another agency will just run more ads and tell you it's "learning".

roichaser

Honestly, I think everyone in this thread is overcomplicating it. Agency owners love to jump in with "clarity on the bottleneck" because it sounds smart and makes them look like the solution. But the real bottleneck is usually that skincare ecommerce is a saturated, low-differentiation space where CTR means nothing if your product isn't solving a genuine pain point. High CTR with terrible CVR? That's not an audience issue, that's a product or pricing problem the ads can't fix.

And the "you're overtesting" advice is a convenient way for agencies to justify charging you for fewer creatives. Ten to fifteen ads at $100/day isn't "thin data" when you've got a clear testing framework-it's standard. The problem is that most agencies don't want to admit their testing process is broken, so they blame the volume.

Also, love the classic "DM me, I can help with an audit" pitch. Really adds credibility to the advice. If you're going to hire anyone, skip the agency and get a freelance strategist who actually works in DTC skincare, not someone who runs a retainer factory. But honestly, you might be better off just burning that money on a better landing page or a stronger offer-those two things will outperform any ad tweak nine times out of ten.

datanerd

a 20-30% CTR with sub 1% conversion indicates a creative-to-audience mismatch. Your landing page isn't the issue - the creative angle promises something the product page doesn't deliver. Match the ad copy's value prop to what's actually on the PDP.

trafficcop

Oh honey, that agency experience? It's basically the industry standard these days. You pay them to nod along and tweak the colour of your CTA button, not to actually think. Meanwhile you're the one doing all the testing and they're just collecting a retainer. Garbage.

Honestly, the problem isn't your ads - it's that you're shouting into Meta's void full of bots and people half-swiping through reels. Your real customers are out there having the exact skincare crisis your product solves, complaining about it in the wild. Reddit, forums, that kind of thing. I've started using a tool that watches those spaces for relevant chatter and hands me a ready-made response. Game changer for getting people who actually need the stuff, instead of bleeding budget on cold traffic.

Forget the agency. Find the desperate people with the itchy, flaky faces and talk to them directly. That's where the conversions live.

datanerd

building out your own analytics layer is the right move, especially when you have an engineer in the founding team. we did something similar for our B2B pipeline - Claude + a quick Python script to pull campaign-level metrics and stitch attribution data into a single dashboard. Saves the agency retainer for what actually needs human creativity.

For skincare ecommerce, burn your budget on proper UTM tagging and server-side tracking before you even think about an agency. Most agencies just run lookalikes on pixel data and call it a day. if you've got the internal chops to build your own dashboards, you'll see way more signal than any black-box agency report. hire a freelancer for creative only, not the whole funnel.