Social Media Marketing ROI: How to Measure What Your Spend Is Actually Returning

Most businesses running social media can tell you their follower count, their average post reach, and roughly how many likes their last campaign got. What they cannot tell you is how much revenue those activities produced. That is the gap this article addresses.

Social media ROI is measurable. It is harder to measure than paid search, and the numbers are messier, but the problem is almost never that measurement is impossible. It is that most businesses are tracking the wrong things.

Why most businesses cannot measure social ROI

The core issue is vanity metrics. Reach, impressions, follower growth, and engagement rate all feel like performance data. They look good in reports. They do not tell you whether social is contributing to revenue.

There are three structural reasons measurement breaks down.

First, the attribution window problem. Someone sees your Instagram post on Tuesday, does nothing, searches your brand name on Thursday, clicks a Google ad on Friday, and converts. Your social team gets no credit. Your Google Ads team gets full credit. In reality, social influenced that conversion, but last-click attribution cannot see it.

Second, platform inflation. Meta reports conversions using a 7-day click, 1-day view attribution window by default. That means if someone saw your ad and then bought something within a week — even if they never clicked — Meta counts it as a conversion. LinkedIn, Google, and TikTok all use different windows. When you add up platform-reported conversions, you typically get 2x to 3x your actual sales volume. The numbers do not reconcile because they were never designed to.

Third, the organic/paid boundary. Organic social affects how buyers perceive your brand, whether they search for you, and whether they trust you enough to click a paid ad. That influence is real, but it does not show up as a trackable click. Businesses without a clear framework end up crediting paid for everything and treating organic as decorative.

The metrics that connect to revenue

Stop reporting these as primary KPIs: reach, impressions, follower count, likes, shares, engagement rate. These are inputs, not outcomes.

Track these instead:

Cost per lead (CPL) — what it costs to generate one enquiry, form fill, or qualified contact from social. Compare across channels.

Lead-to-opportunity rate — of the leads social generates, how many become actual sales conversations. A low rate here means either the audience quality is wrong or the offer is attracting the wrong people.

Cost per acquisition (CPA) — what it costs to generate one paying customer from social. This is your primary performance number for direct response campaigns.

Revenue attributed per channel — using CRM data and UTM tracking, how much closed revenue traces back to each social channel. This requires proper setup, but it is the only number that answers the actual question.

ROAS (return on ad spend) — for e-commerce and direct purchase campaigns. Revenue divided by ad spend. A 3x ROAS means for every £1 spent, you generated £3 in sales.

Setting up a measurement framework

You need three things in place before any social ROI number is trustworthy.

UTM parameters on every link. Every social post, ad, and bio link should carry UTM tags (source, medium, campaign, content). Without these, GA4 cannot tell you which social channel or campaign sent the visitor. This is basic, and a lot of businesses skip it or apply it inconsistently.

A CRM with lead source tracking. When a form is filled in, the lead source should be captured and stored. If your CRM does not record where a lead came from, you cannot connect social activity to pipeline and closed revenue. This matters most for B2B, where the conversion event (a demo request) and the revenue event (a signed contract) are separated by weeks or months.

GA4 with conversion events properly defined. Not pageviews — actual conversion events. Form submissions, phone click-throughs, purchase completions, checkout starts. Configure these as key events in GA4 and import them into your ad platforms for bidding.

Once these three are working together, you can start pulling honest numbers from your campaigns.

Attribution models for social media

The attribution model you choose changes the ROI number you get, sometimes by 50% or more.

Last-click attribution assigns all credit to the final touchpoint before conversion. It systematically undercounts social, particularly organic social and top-of-funnel paid, because social usually appears earlier in the journey.

First-click attribution overcounts social in the opposite direction, giving it full credit even when seven other touchpoints happened after it.

Data-driven attribution, available in GA4 and Google Ads, uses your own conversion data to distribute credit across touchpoints based on actual patterns. It is the most accurate model for most businesses, provided you have enough conversion volume (roughly 300 to 500 conversions per month is the threshold where it becomes reliable).

For B2B businesses with long sales cycles, the honest answer is that no platform model is sufficient. You need CRM-based attribution where a salesperson or your marketing automation tool logs the original lead source against the deal. When a contract closes six months after a LinkedIn ad impression, only your CRM can connect those dots.

Platform-by-platform ROI benchmarks

These are working benchmarks based on typical performance ranges. Your numbers will vary based on industry, offer quality, and how well your funnel is set up.

Instagram and Facebook: where most consumer ROI lives

Meta remains the most measurable platform for direct response because the ad infrastructure is mature and the targeting options are specific. E-commerce brands running retargeting campaigns — ads targeting people who visited the site, added to cart, or purchased before — regularly see 4x to 8x ROAS. Cold prospecting campaigns to new audiences typically run 1.5x to 3x, which works if your margins support it.

The CPL for lead generation on Meta varies widely by industry. Financial services and insurance leads often cost £30 to £80. Home improvement and trade services range from £15 to £50. Consumer service offers can come in below £10 if the audience is large and the offer is strong.

The biggest lever on Meta ROI is creative. Audience targeting has narrowed as iOS privacy changes reduced signal quality. The ad itself — the image, video, or copy — now does more of the targeting work, because good creative self-selects the right audience. Businesses that refresh creative every 3 to 4 weeks consistently outperform those running the same ads for months.

LinkedIn: expensive per click, but the right audience

LinkedIn CPCs average £5 to £12 for most B2B audiences. That is 4x to 8x more expensive than Meta. The reason it still works for B2B is audience precision. You can target by job title, seniority, company size, and industry in a way no other platform matches.

A £120 CPL looks terrible until you consider that the lead is a CFO at a 300-person company with the authority to sign a £50,000 contract. The ROI calculation is different from B2C entirely.

The biggest mistake on LinkedIn is treating it like a direct response channel. Most LinkedIn campaigns that fail are asking cold audiences to book a demo or request a quote in the first impression. The platform works better as a two-step process: first, offer something useful (a guide, a benchmark report, a webinar) to generate a lead at reasonable CPL. Then retarget those leads with your sales offer.

YouTube: harder to measure, harder to ignore

YouTube ROI is real but difficult to track cleanly. View-through conversions — where someone watches your ad and later converts without clicking — are genuine but hard to attribute confidently. Businesses that import Google Ads conversion data from GA4 and use view-through attribution get a clearer picture than those relying on platform-reported numbers alone.

The strongest YouTube ROI tends to come from mid-funnel campaigns targeting warm audiences: people who visited your website, watched a previous video, or are on your customer list. For these audiences, YouTube acts as a reminder and trust signal rather than a cold introduction.

When to increase vs. cut social spend

Increase spend when: CPA has been stable or improving for at least 4 weeks, conversion volume is above 50 per month per campaign (enough data to be statistically meaningful), and lead quality matches your sales team's expectations.

Cut spend when: CPA has risen more than 20% for two consecutive weeks without a clear explanation, lead-to-opportunity rate drops below your historical baseline, or you cannot attribute any closed revenue to the channel after 90 days of proper tracking.

The grey area is organic social. The answer there is almost never to cut it entirely, because it affects brand search volume and paid ad performance indirectly. The question is whether the time investment is proportionate to the audience size and engagement you are getting back.

Seasonal variations matter too. Most social platforms see cost increases of 20 to 40% during Q4 as retail advertisers flood the auction. If your business does not benefit from holiday spending, scaling back paid social in November and December and reinvesting in January is often a better use of budget.

What realistic social ROI looks like by industry

E-commerce — total ROAS (organic plus paid combined) of 3x to 5x is achievable with good creative and a retargeting structure. Below 2x is a signal that either the margins are too thin or the funnel has gaps.

B2B professional services — pipeline-to-spend ratio of 5x to 10x over a 6-month window is reasonable. Shorter windows under-attribute because deals take time to close.

SaaS — cost per free trial or demo request from social tends to run £30 to £120 depending on the product price point. What matters is trial-to-paid conversion rate, which varies so much by product that benchmarks are almost useless here.

Local services (trades, clinics, legal) — Meta drives the strongest ROI for most local service businesses. A well-run local campaign with a £2,000 monthly budget can generate 60 to 120 leads per month in a mid-sized city, at CPLs of £15 to £35.

If you want to understand how a properly structured social media programme connects to revenue for your business type, our social media marketing service covers strategy, paid campaigns, and the tracking setup needed to measure what matters.

The short version

Vanity metrics are easy to report and useless for decisions. The metrics worth tracking are CPL, CPA, lead-to-opportunity rate, and revenue attributed per channel. You need UTM parameters, CRM lead source tracking, and GA4 conversion events before any ROI number is trustworthy. Attribution models matter — data-driven is the most honest for most businesses, and CRM-based attribution is the only option for long B2B sales cycles. Social ROI varies sharply by platform and industry, and the decision to scale or cut should be based on CPA stability over at least 4 weeks of data, not platform-reported vanity numbers.