Linked Media

Home
Linked Media

Here at Linked Media, we are a group of talented performance marketers in the e-commerce & info-product space.

We pride ourselves in working with industry leading clients, scaling them aggressively and profitably.

28/08/2026

Doubling the budget is not scaling. It's a reset.

The scenario every e-com founder knows: an ad set is flying at 4X. Obvious move — take it from £200/day to £500/day and enjoy the weekend.

Monday morning it's at 1.4X and nobody can explain what happened.

Here's what happened. A significant budget change pushes the ad set back into the learning phase. It re-explores, delivery widens, CPMs shift, and the optimisation profile you spent three weeks and real money building gets partially thrown away.

You didn't scale a winner. You reset it and then judged it on its worst week.

How to scale without breaking it:

→ Increase by 20–30% at a time, then hold for 48–72 hours
→ Only increase again once performance stabilises at the new level
→ For bigger jumps, duplicate the ad set at the higher budget and let it learn in parallel — the original keeps earning
→ Never increase budget and change creative in the same move. You'll never know which one caused what
→ Scale on a 7-day rolling average, not on yesterday

Slower is faster. A 20% increase every three days is 6X the budget in a month, with the learning intact.

27/08/2026

One ad. One idea. Or you learn nothing.

Here's a test we see constantly: three "variants" launched together. Different hook, different offer, different format, different music, different edit.

Ad B wins. Brilliant. What did you learn?

Nothing. You have no idea which of the five changes did it, so you can't apply it to the next batch. You bought a winner instead of buying knowledge, and winners fatigue in six weeks.

How to structure a test that actually teaches you something:

→ Change ONE variable per ad. Hook, or offer, or format. Never all three
→ Same audience, same placement, same budget
→ Give it 3–5 days and at least 50 conversions before you call it. Day-two results are noise
→ Write the hypothesis down BEFORE you launch. "I think a problem-first hook beats a product-first hook." Then you find out whether you understand your customer
→ Log the result. Win or lose. Build the library

Do this for three months and you stop guessing. You'll have a documented list of what your specific audience responds to, and every new batch starts from a higher floor.

Buying winners is a treadmill. Buying knowledge compounds.

26/08/2026

Three numbers tell you exactly where your funnel is broken. Ten minutes, no analyst required.

CTR → the creative problem
Below 1%? People are seeing the ad and choosing not to engage. The issue is upstream of everything else. Do not touch the landing page.

CVR → the page problem
CTR healthy but site conversion under 1.5%? You're paying for the right people and losing them on arrival. Page speed, message match, trust signals, checkout friction.

AOV → the offer problem
CTR and CVR both fine but you can't scale profitably? You don't have a marketing problem. Your basket is too small to support the CPA. Bundles, tiered discounts, post-purchase upsells.

The discipline is in what you do next.

Whichever one is below benchmark is the ONLY thing you work on this week. Not all three. Not "a bit of everything".

Most brands lose entire quarters by optimising the two things that weren't broken, then concluding that paid social doesn't work for them.

Diagnose. Fix one. Re-measure. Repeat.

24/08/2026

Your ad promised one thing. Your landing page says another.

It's the cheapest conversion rate win in e-commerce and it's sitting unclaimed in most accounts right now.

Do this today. Open your best-spending ad on your phone. Click it like a customer would. Then look at the first screen of the page without scrolling.

Ask three questions:

1. Does the page repeat the exact promise the ad made? Same words. Not a paraphrase, not a "brand-appropriate" rewrite
2. Is the product from the ad the first thing visible? Or did you land on a category page and now they're hunting
3. Does it look like the same company? Same imagery, same model, same colours

If the ad said "20% off your first order" and the page says "New Arrivals", you have just paid for a click and then asked the customer to do the work of connecting them. Most won't.

The fixes are genuinely small:

→ Mirror the ad headline as the page H1
→ Send traffic to the product, not the collection
→ Use the exact creative from the ad as the hero image

We've seen this alone move CVR 20–40% with no change to the ad, the audience or the budget.

21/08/2026

14 ad sets. 3 conversions each. Zero learning.

This is the most common structural problem we see in £15k–£100k/month accounts, and it's usually the result of good intentions.

Someone wanted to test properly. So they built an ad set per interest, per lookalike, per age band. Now the budget is spread across 14 ad sets, each getting a trickle, none of them ever escaping the learning phase.

Meta needs roughly 50 conversions per ad set per week to optimise properly. Below that, delivery is basically guessing.

Do this maths on your own account:

Weekly budget ÷ number of ad sets ÷ your CPA = conversions per ad set per week

If that number is under 50, you are paying to learn nothing. Fourteen times over.

The fix:

→ Consolidate down to 2–3 ad sets, minimum
→ Let broad targeting do the work. The algorithm finds the audience better than your interest stack does
→ Test at the creative level, not the audience level
→ Only split an ad set out when it can carry 50+ conversions a week on its own

Fewer, bigger, cleaner. Almost every account gets better the week after this.

19/08/2026

ROAS tells you if the ad works. MER tells you if the business works.

You need both. Most brands track one and wonder why the numbers never reconcile.

ROAS — revenue attributed to an ad ÷ spend on that ad.
Use it to judge creative, ad sets and campaigns against each other. It's your editing tool.

MER — total business revenue ÷ total ad spend (everything, every platform).
Use it to judge whether the machine as a whole is making money. It's your steering wheel.

Why you need the pair:

You can run a 5X ROAS campaign and still shrink, if it's 8% of your revenue and the rest of the account is bleeding.

You can run a 2.6X ROAS campaign that lifts blended revenue across the whole business, because it's feeding brand search, email signups and repeat orders.

Simple rule we use:

→ Judge individual creative on ROAS
→ Judge scaling decisions on MER
→ If ROAS climbs while MER falls, you're not scaling. You're reallocating

Two numbers, one dashboard. Check them side by side every Monday.

17/08/2026

Run this one report before you kill another ad.

In Ads Manager, switch your attribution setting from 7-day click to 1-day click and look at the same date range twice.

You'll see two very different businesses.

7-day click: £48,200
1-day click: £21,700

That gap is the most important number in your account and almost nobody looks at it.

How to read it:

Small gap (under 30%) — your ads are driving fast, decisive purchases. Scale with confidence.

Huge gap (over 60%) — a large share of your "attributed" revenue is people who took a week to buy. Some of them were always going to buy. Meta is taking credit for your email list, your brand search and your returning customers.

What to do with that:

→ Compare per campaign, not just account-wide. Retargeting is usually the worst offender
→ Cross-check against a post-purchase survey — one question, "how did you hear about us?"
→ Watch your blended numbers when you scale. If platform revenue grows and blended doesn't, you found the gap

None of this means Meta is lying to you. It means you should know which version of the truth you're making decisions on.

14/08/2026

If your 3-second hook rate is under 25%, nothing that happens after it matters.

Hook rate = 3-second video plays ÷ impressions.

It's the single fastest diagnostic in your account and almost nobody has it as a column.

What it tells you:

Under 20% — the first frame is invisible. The problem is not your offer, your product or your landing page. Three quarters of people never even saw the ad.

25–35% — healthy. Now the rest of the ad is worth optimising.

Over 40% — your hook is doing real work. Protect it, and reuse that opening on other concepts.

The fix is almost always in the first 1.5 seconds:

→ Open on motion or a face, never a logo
→ Put the product on screen immediately
→ Burn the promise into text on frame one — most people watch on mute
→ Cut the "hi guys, so today I wanted to talk about..." entirely

Add hook rate as a custom column in Ads Manager this afternoon. It reorders your whole priority list.

12/08/2026

You don't have a budget problem. You have a creative volume problem.

The brands that plateau at £15k–£50k a month almost always have the same account: three ads that worked in March, still running in August, with the budget nudged up every few weeks and the CPA quietly climbing.

Here's the uncomfortable maths.

Roughly 1 in 6 new creatives beats your current control. Sometimes 1 in 10.

So if you ship 3 new ads a month, you're statistically buying yourself one winner every other month. Meanwhile your existing winners are fatiguing at 4–6 weeks.

You are losing creative faster than you're replacing it. That's the whole plateau.

What to do instead:

→ Ship 10–12 new concepts a month, not 3
→ Concepts, not colour variations. A new headline isn't a new ad
→ Batch produce. One shoot day, one editing day, twelve assets
→ Keep a running "graveyard" doc of what failed and why

Budget doesn't create demand. Creative does. Budget just buys more of whatever your creative already earned.

11/08/2026

Most brands scale to a number they have never actually calculated.

Ask a founder what ROAS they need and you usually get a feeling, not a figure. "About 3?" "Whatever we did last month?"

Here is the whole calculation:

1 ÷ gross margin = your break-even ROAS

→ 40% margin = 2.5X break-even
→ 30% margin = 3.33X
→ 60% margin = 1.67X

Gross margin means after COGS, shipping, payment fees and pick/pack. Not the number on your P&L that flatters you.

Now go and look at your account. Every ad set sitting below that figure is not "underperforming". It is losing you money on every single order it brings in.

And the ones sitting above it that you paused because "the ROAS looked low"? You paused a profitable ad.

Work out your number tonight. Write it on a post-it. Stick it to the monitor. It is the only threshold that matters when you're deciding what to cut.

Address

Opening Hours

Monday	11:00 - 15:30
Tuesday	11:00 - 15:30
Wednesday	11:00 - 15:30
Thursday	11:00 - 15:30
Friday	11:00 - 15:30
Saturday	11:00 - 15:30
Sunday	11:00 - 15:30

Website

http://www.linkedmedia.co.uk/

Alerts

Be the first to know and let us send you an email when Linked Media posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Contact The Business

Send a message to Linked Media:

Want your business to be the top-listed advertising & marketing company?

Country:

City: