Paid Social Advertising
The ad is the part you see. It’s usually not the part that decides.
When paid social underperforms, the ad gets blamed. More often the constraint is somewhere else — who it reached, what it asked for, what happened after the click, or how long the lead sat before anyone called. The pieces have to reinforce one another, and a weak link can cap everything else.
The operating model
Five parts, and they depend on each other.
Paid social isn’t a single lever. Five things have to work together, and strong performance requires them to reinforce one another rather than compensate for one another. That’s why “the ads aren’t working” so often turns out to be a diagnosis of something that isn’t the ads.
Factor 01
Audience
Who the campaign reaches, and who it deliberately excludes.
Factor 02
Creative
Whether it earns attention in a feed built to be scrolled past.
Factor 03
Offer
Whether there’s a clear reason to act, and act now.
Factor 04
After the click
The page or form that has to keep the promise the ad made.
Factor 05
Follow-up
What happens to the enquiry once it exists.
What a constrained campaign looks like
Four parts working. One holding the rest back.
The account looks healthy in the platform and disappointing in the business.
Four parts are doing their job. The fifth is where the value leaks out — and no amount of additional budget on the first four recovers it.This is an illustrative operating model, not a performance formula. We’re not suggesting a single imperfect part reduces results to nothing, or that any of this can be calculated in advance. The point is narrower and more useful: a serious weakness in one part can heavily constrain the value the other four create, which is why diagnosis comes before spending more.
Audience & retargeting
The best audiences are usually built from people who already bought.
When enough good first-party data exists and platform rules allow its use, it becomes some of the most valuable material available — for building audiences, setting exclusions, seeding lookalikes and feeding optimisation. It’s information you already own, and most accounts never connect it.
What you already have
Source
Past converters
Your customer list. The smallest group and by far the most useful, because they’ve already proved they buy.
Source
Site visitors
Everyone who landed on your site, captured by the pixel. Larger, mixed intent, still far better than a guess.
Source
Near misses
People who viewed a product, started a form, or reached checkout and stopped. The warmest group you’re not currently reaching.
What we build from it
Built
Lookalike audiences
New people who resemble your actual buyers on the signals the platform can see. Reach beyond your followers, without abandoning targeting.
Built
Exclusions
Existing customers removed so you stop paying to advertise to people who already bought. Quietly one of the biggest sources of waste.
Built
Remarketing
Campaigns aimed squarely at the near misses — people who already showed interest and stopped. Frequently the group nobody is currently reaching.
If you don’t have traffic or a customer list yet, we start with interest-based targeting and build the data as the campaigns run — but we’ll tell you plainly that the first months are constructing an asset, not harvesting one.
How targeting actually works now
Manual audience selection is one lever among several. Modern paid-social platforms combine what the advertiser specifies with automated delivery, creative signals and conversion data to decide who actually sees an ad.
That means broad targeting sometimes outperforms tightly constrained targeting, and sometimes the reverse is true. Which approach fits depends on your business model, audience, platform, conversion volume, geography, category restrictions and how much first-party data you have. We test rather than assume, and we’d rather say that than sell you precision we can’t demonstrate.
Retargeting reaches a different decision state
Someone who has already visited your site, engaged with the brand, watched content or started a form may be at a different point in their decision than a completely cold prospect. Retargeting can support familiarity, return visits, decision reinforcement, reminders and offer follow-up. For online stores, a connected product catalog lets remarketing show the specific products someone viewed rather than a generic ad.
It doesn’t guarantee conversion, and what’s available depends on the platform, the category and current privacy and policy constraints. If you’re advertising in healthcare or another sensitive category, the tracking behind all of this needs particular care — see HIPAA-sensitive conversion tracking.
Creative & offer
The two that decide whether anyone stops scrolling.
The audience can be right and the campaign can still struggle, because a feed is a place people move through quickly. These two factors decide whether the impression becomes anything at all.
What a creative test should actually compare
Same offer in all six versions — a free in-home estimate. Each ad has a hook, a visual, a message and a call to action. Outlined is what changed.
Three versions that differ only in the button.
What it can teach you: which button label gets slightly more clicks from people who already stopped scrolling.
Three different reasons someone might care, same offer underneath.
What it can teach you: which reason makes this audience stop at all — and that shapes the next round of creative, the landing page and the offer itself.
Creative
Not the polish. The hook, the visual, the first line, the format, and whether the message fits the person seeing it.
- Hook and opening line
- Visual and format
- Headline and body copy
- Call to action
- Proof and credibility signals
- Fit between message and market
Creative can lose effectiveness as an audience sees it repeatedly. How quickly depends on audience size, spend, frequency, the creative itself, the platform, the category and the objective — which is why creative development is an ongoing task rather than only a launch one.
Offer
What the person is actually being asked to do, and whether it’s worth doing. Worth answering honestly before spending:
- Why should this person care?
- Why should they act now rather than later?
- What exactly are they being asked to do?
- Is the value clear without explanation?
- Does it fit this audience and this platform?
- Is there enough proof to make it believable?
- Is it asking for too much commitment too early?
Paid social can distribute an offer very effectively. It can’t make an unconvincing one compelling through targeting alone.
Paid distribution cannot indefinitely compensate for weak market demand, unclear positioning, or an offer that isn’t landing. When that’s the constraint, more budget makes the problem more expensive rather than smaller.
Creative scope covers image and text creative, ad copy, headlines, URLs, calls to action, and testing variations. Finished video you supply can be used. Professional video production, filming, creator sourcing and influencer work aren’t part of standard scope and would be quoted separately.
After the click
Where campaigns quietly lose most of what they earned.
The ad’s job ends at the click. Everything after it belongs to the conversion path, and this is the factor most often left untouched while budget gets increased instead.
Option A
Native lead forms
Potential advantages- Lower friction, fewer steps
- Faster completion, often on mobile
- Stays inside the platform experience
- Some fields can prefill
- Less room for context and proof
- Less control over the experience
- Qualification may need extra questions or follow-up design
Option B
Landing pages
Potential advantages- Room for context, proof and branding
- More control over the conversion experience
- More flexibility in how you qualify
- Easier to test structurally
- Another page load between ad and action
- Potentially more friction
- Website performance becomes part of the result
Neither is universally better. Lead quality depends on the offer, targeting, form design, the questions asked, the audience, the follow-up and the category — not on which mechanism you picked. We choose per campaign, and sometimes test both.
What commonly constrains this stage
- Slow mobile experience
- Confusing page
- Unnecessarily long form
- Weak trust signals
- Ad-to-page mismatch
- Unclear next step
- Broken tracking
Not every campaign needs a purpose-built landing page. But where the destination is what’s limiting results, that’s a web development problem rather than a bidding one, and spending more on media won’t fix it.
And then there’s follow-up
A lead is an opportunity, not a completed sale. Contact speed, persistence, qualification and how the conversation is handled can all materially influence contact rate, booking rate, close rate, and ultimately what a customer costs to acquire.
The longer an interested lead sits untouched, the more opportunity can be lost — though the right response process genuinely depends on your business and the type of lead. What’s worth saying plainly is the scope boundary: we manage the campaign. Sales representatives, call handling, appointment setting and CRM nurture sequences aren’t part of this service unless separately scoped. For the deeper argument about what a lead actually costs once it becomes a customer, see paid lead generation.
How it runs
Tracking goes in before any money goes out.
This ordering isn’t a preference. Spend that happens before tracking exists can never be evaluated afterward — the data simply isn’t there to look at.
Objective, audience & data
What the campaign is actually for, who it needs to reach, and what first-party data already exists to build from — customer lists, site traffic, past converters. This decides whether we’re harvesting an audience asset or constructing one.
Tracking & campaign architecture
Ad accounts, business managers and page access connected, then pixel and conversion tracking implemented and verified. Campaign structure built so results can be read by audience and creative rather than as one undifferentiated number. Nothing meaningful gets spent until this works.
Creative, offer & conversion path
Ad copy, images, headlines and calls to action built and set up to be compared against each other. The offer gets pressure-tested before launch, and the destination — form or landing page — gets checked rather than assumed. If you have pre-approved brand assets, handing those over speeds this up considerably.
Launch, measurement & iteration
Remarketing brought online, creative rotated as performance shifts, audiences and exclusions maintained, and results read against business outcomes rather than platform metrics alone. Where you can share what happened to the leads, that feeds back into the next round.
Tracking first
No campaigns without conversion tracking
If it can’t be installed on your site, we’ll say so before taking the work rather than spending your budget blind.
3 months minimum
Enough time to learn something
Setup, creative iteration and enough conversion data to optimise against. Judging paid social in week three is judging it mid-experiment — though a three-month commitment isn’t a promise that any campaign will be profitable within it.
Measured outcome
More of what the business wanted, on less spend.
Paid social — recruiting
254.54%
More applications, from 44 to 156
A recruiting campaign that produced more while spending less
Applications are the outcome that mattered to this business, and they rose substantially while the monthly media budget came down. That’s the shape of result this page is arguing for — a business outcome moving, not a platform metric.
We can report what the campaign produced. We don’t have evidence isolating how much came from targeting, creative, offer, campaign structure or optimisation individually — so it’s presented as a paid social campaign outcome rather than credited to any one factor. One campaign is evidence of what happened in that engagement, not a guaranteed result for another advertiser.
Measurement
Platform metrics explain the campaign. Business metrics decide whether it’s worth running.
Both matter, and they answer different questions. Reporting that only shows the first kind is reporting that avoids the second.
Diagnostic
How the campaign is behaving
- Impressions and reach
- CPM
- Click-through rate
- Cost per click
- Frequency
Useful for diagnosis. Click-through rate says something about creative; cost per click says something about efficiency. Neither is a business result on its own.
Business
Whether it’s producing anything
- Lead volume
- Cost per lead
- Conversion rate
- Cost per qualified lead
- Booked calls, appointments or applications
- Purchases, for e-commerce
- Customer acquisition cost, where trackable
Not every business can track every one of these. What’s measurable depends on your systems, and we’ll tell you which parts of the chain we can actually see.
On attribution
Platform-reported conversions are attributed according to that platform’s measurement methodology. View-through attribution, cross-device behaviour, privacy and browser restrictions, delayed conversions, offline sales and multiple touchpoints all mean those numbers may not reconcile exactly with your CRM or financial records.
That doesn’t make the data useless — it’s genuinely useful attribution data with methodological limits, and we report it as such rather than presenting it as a complete picture.
Straight answer
Four things we won’t do with your ad account.
All four make paid social cheaper to deliver. All four are why so much of it produces impressions and very little else.
01
Run campaigns without conversion tracking
Without it, nobody can tell which ad produced the outcome. The reporting looks fine and explains nothing.
02
Leave creative untouched indefinitely
Audiences see the same ad repeatedly and respond less over time. Performance can decay quietly while spend stays flat.
03
Report success from impressions alone
Reach is not the product. If the reporting can’t connect to something the business cares about, it isn’t reporting.
04
Keep spending against a broken path
If the offer or the conversion experience is what’s failing, more budget makes it more expensive. We’ll say so rather than quietly scaling it.
Where these run
Platform follows the audience and the objective.
Meta is often relevant for consumer, local-service, lead-generation, recruiting, retargeting and e-commerce work — though not universally the right primary platform. LinkedIn can matter for B2B, professional audiences and longer, higher-value sales cycles. TikTok, Pinterest and X fit particular categories rather than most. We’d rather run two platforms properly than six thinly.
- TikTok
- X
Google, Microsoft and YouTube advertising is handled separately under Paid Advertising — those reach people actively searching, which is a different discipline. Content and community without ad spend is Organic Social. A combined approach using both paid distribution and ongoing organic presence is also possible — the law firm engagement was run that way. Industry shapes the mix; see industries we work in. For online stores, Shopping campaigns sit under e-commerce paid advertising, and both are part of our wider e-commerce marketing work.
Fit check
Who this works for.
A good fit
- You have a clear offer people can understand quickly
- You have site traffic or a customer list to build audiences from
- You can get conversion tracking installed and keep it working
- Someone follows up on enquiries reliably
- You’re willing to test creative and messaging rather than run one ad
- You’ll share what happened to the leads downstream
- You judge results on outcomes, not reach
Probably not
- You expect guaranteed sales, quickly
- Nobody is going to follow up on the leads
- There isn’t a meaningful offer yet
- You can’t take on more demand right now
- You expect one ad to run indefinitely
- Followers and impressions are what you want measured
- Tracking can’t be installed, for whatever reason
Questions
Paid social FAQs.
What exactly is a lookalike audience?
You give the platform a list of people you already know are good — past customers, or visitors who converted. It finds other users who resemble them on the signals it can see, and builds a new audience from those people.
It’s one of the more effective ways to reach people who don’t know you yet, and it’s only as good as the list you seed it with. Availability varies by platform, category and region.
Do you set up the conversion tracking?
Yes, as part of onboarding and before meaningful spend begins — pixel and analytics conversion tracking both. If tracking genuinely can’t be installed on your site, this service isn’t the right fit and we’ll tell you rather than proceeding anyway.
Should we use native lead forms or send traffic to a landing page?
It depends on the campaign, and we’d be suspicious of anyone who answers that without asking questions first. Native forms reduce friction; landing pages allow more context and control. Which produces better outcomes depends on your offer, your audience, how the form is designed, what you ask for, and how quickly you follow up. Sometimes we test both.
Is there a minimum ad spend?
It depends on your goals and how large your audiences are. We scope the right starting budget together during onboarding rather than applying one number to every business. A budget too small to gather data isn’t a cheap test, it’s a slow way to learn nothing.
For online stores, we set out the full method in how much an e-commerce business should spend on marketing.
How do you pick which platforms to use?
Where your buyers actually spend time, what your offer suits, what your budget supports, and what you’re trying to achieve. Category restrictions and targeting availability matter too. If a platform you’re interested in isn’t a sensible fit yet, we’ll say so plainly instead of adding it to widen the scope.
We run an online store. Should we start with Meta or Google?
It depends mostly on the product. Google tends to suit products people already search for; Meta tends to suit products that benefit from being seen and demonstrated. Plenty of stores eventually run both, but on a limited budget it’s usually better to fund one properly than both thinly. Meta Ads vs. Google Ads for e-commerce walks through how to choose.
In a real e-commerce analysis we published, about $20 a day was spread across roughly 15 Meta ads — too little behind any one of them to learn much. Concentration mattered more than which platform was chosen.
How is this different from organic social?
Organic social is content and community management with no ad spend behind it. This is paid advertising: audience development, tracking, creative and budget aimed at reaching people who don’t know you yet. They complement each other, and the social media hub explains how they behave differently over time.
How quickly will we see results?
Activity comes quickly — impressions and clicks arrive as soon as campaigns run. Reliable performance takes longer, because early spend is buying data as much as buying results. The first weeks go to testing creative, letting audiences build and accumulating enough conversions to optimise against. We won’t put a date on profitability, because that depends on factors outside the ad account.
Next step
Let’s see what data you’re already sitting on.
Most businesses have more to build audiences from than they realise — a customer list, site traffic, people who nearly bought. One conversation is enough to work out whether there’s enough there to make paid social worth running, and which of the five factors is most likely to be your constraint.