Social Media ROI: Stop Measuring the Wrong Things

Yes, you can measure the ROI of social media marketing, but most businesses are measuring the wrong things and drawing the wrong conclusions. The real problem is not a lack of data. It is that the metrics most social platforms surface are designed to justify continued ad spend, not to tell you whether your marketing is actually growing the business.

Get the measurement framework right, and the strategic questions answer themselves. Get it wrong, and you will spend years optimising for numbers that have no relationship to revenue.

I have spent twenty years watching businesses confuse measurement with accountability. Early in my agency career, I was as guilty as anyone of presenting engagement reports that looked impressive and meant very little. A client would ask whether their social spend was working, and we would show them reach figures, follower growth, and click-through rates. Nobody in the room, including me, could honestly connect those numbers to the P&L. That changed when I started running agencies with full commercial responsibility. When you own the revenue number, you stop tolerating metrics that do not point toward it.

Why Social Media ROI Is Genuinely Difficult to Measure

Social media sits across multiple stages of the customer experience simultaneously. A single platform can drive initial awareness, sustain consideration, and close a conversion, sometimes in the same session, sometimes across weeks. That makes clean attribution structurally difficult in a way that, say, paid search is not.

Paid search captures existing intent. Someone types a query, clicks an ad, and buys. The causal chain is short and visible. Social media, particularly organic social, often works earlier in the process. It shapes the preference that eventually becomes the search query. By the time the attribution tool fires, social’s contribution has already happened and gone unrecorded.

This is not a reason to abandon measurement. It is a reason to build a more honest model of what you are actually trying to measure. The case for social media marketing has always rested on more than last-click conversions, and measurement frameworks need to reflect that reality.

If you want a broader grounding in how social fits into a coherent channel strategy, the Social Growth and Content hub covers the full picture, from platform selection to content strategy to paid social.

The Attribution Problem Most Marketers Ignore

Platform-reported attribution is not neutral. Meta, LinkedIn, TikTok, and every other major social platform have a commercial interest in showing you that their platform drove your results. Their attribution windows are set generously. View-through attribution, where a conversion is credited to an ad someone saw but never clicked, can be set to 28 days on some platforms. That means someone who saw your ad once and then bought something four weeks later through a completely different channel gets counted as a social conversion.

I ran a test on this years ago with a mid-sized retail client. We compared platform-reported conversions against what we could see in their analytics and their actual revenue data. The gap was significant. Platform attribution claimed roughly three times the conversions that could be verified through any other measurement method. Nobody had been deliberately misled. The platform was simply using its own attribution rules, and we had never questioned them.

The lesson is not that social did not contribute. It almost certainly did. The lesson is that platform-reported numbers are a starting point for analysis, not a conclusion. Treat them as one signal among several, not as ground truth.

A more reliable approach layers multiple measurement methods. Analytics platform data, CRM pipeline data, revenue data, and brand tracking surveys all give you different angles on the same question. When they point in the same direction, you have something worth acting on. When they diverge, you have something worth investigating.

What Incrementality Testing Actually Tells You

Incrementality is the question that most social measurement frameworks never ask: how much of this result would have happened without the social activity?

I spent a long time earlier in my career overvaluing lower-funnel performance marketing, and social was no exception. The numbers looked good. Conversions were attributed, costs per acquisition were within target, and the channel appeared to be delivering. What I was slower to recognise was that a meaningful portion of those conversions were people who were already in market and would have found the product through another route. Social had not created the demand. It had intercepted it, and intercepting demand is worth something, but it is worth considerably less than creating it.

The clothing retail analogy I keep coming back to: someone who has already tried on a jacket in a physical store is far more likely to buy it than someone browsing the website cold. If your retargeting ad reaches that person and they convert, the ad gets the credit. But they were already most of the way there. The incrementality of the ad, what it actually added to the probability of purchase, may be quite small.

Incrementality testing, running holdout groups who are deliberately excluded from seeing your social activity and comparing their behaviour against the exposed group, is the most reliable way to answer this question. It is not simple to run well, but the insight it produces is worth the effort. If your holdout group converts at nearly the same rate as your exposed group, you are spending money to claim credit for things that would have happened anyway.

Platforms including Meta have their own incrementality testing tools. They are imperfect, but they are better than relying on standard attribution reporting alone. A structured approach to social media strategy should include measurement methodology from the outset, not bolted on after the campaign has launched.

How to Measure Organic Social ROI Without Losing Your Mind

Organic social is harder to quantify than paid, but that does not mean it is unmeasurable. It means you need different metrics and a longer time horizon.

The mistake most teams make is applying paid social logic to organic. They look for direct conversions attributable to individual posts and, finding very few, conclude that organic is not working. That is the wrong framework. Organic social builds familiarity, credibility, and preference over time. Those are real commercial outcomes. They just do not show up in last-click reports.

Useful proxies for organic social ROI include share of voice relative to competitors, branded search volume over time, direct traffic trends, and the quality of inbound leads, specifically whether people who mention your social content in sales conversations convert at a higher rate. None of these are perfect. Together, they give you a defensible picture of whether organic social is doing useful work.

Content quality is also a factor that measurement often ignores. How you approach content creation directly affects the signals you can measure. Content that generates genuine conversation and saves or shares is doing different work than content that generates passive impressions. Track the former more carefully than the latter.

If you are running organic social at any meaningful scale, a structured content calendar is not optional. Social media calendar tools help you maintain consistency and create the longitudinal data you need to spot trends. You cannot measure what you have not tracked systematically over time.

Building a Measurement Framework That Connects to Business Outcomes

The measurement frameworks I have seen work best start from the business outcome and work backward, rather than starting from available platform metrics and working forward. The difference sounds subtle. In practice, it changes everything.

Start with the commercial question you are trying to answer. Are you trying to grow new customer acquisition? Reduce cost per acquisition? Increase customer lifetime value? Grow brand preference in a specific segment? Each of these requires a different measurement approach and a different set of leading indicators.

When I was growing an agency from around 20 people to over 100, one of the things that changed most significantly was how we talked about measurement with clients. Early on, we led with platform metrics because they were easy to produce and looked impressive. As the agency matured, we pushed harder to connect our activity to things clients actually cared about: revenue, pipeline, customer acquisition cost, market share. That shift was uncomfortable at first. It meant holding ourselves to a higher standard of accountability. It also meant clients trusted us more and stayed longer.

A practical framework for social ROI connects three levels of measurement. Platform metrics (reach, engagement, click-through) are your operational layer. They tell you whether your content and targeting are working mechanically. Business metrics (leads, pipeline, revenue attributed) are your commercial layer. They tell you whether the channel is contributing to growth. Brand metrics (awareness, consideration, net promoter score, share of voice) are your strategic layer. They tell you whether social is building the kind of long-term equity that makes future marketing more efficient.

Most businesses only track the first layer. The best ones track all three and understand how they relate to each other.

The Honest Conversation About Social Media and Brand Building

One of the things I observed judging the Effie Awards is that the campaigns with the strongest measurable business impact almost always had a brand-building component working alongside the performance activity. The purely performance-driven social campaigns, optimised entirely for short-term conversion metrics, tended to produce diminishing returns over time. The brand-building campaigns were harder to evaluate in the short term but showed up clearly in the longer-term business data.

This matters for how you think about social ROI. If you measure only what is easy to measure, you will systematically underinvest in brand-building activity and overinvest in conversion-focused activity. Over time, that erodes the brand equity that makes your performance marketing work. You end up spending more to reach people who are less predisposed to buy from you.

The fix is not to stop measuring performance. It is to acknowledge that social media does work that sits outside the reach of standard attribution tools, and to build measurement approaches that account for that. Brand tracking surveys, share of voice analysis, and marketing mix modelling are all imperfect tools. They are also considerably more honest than pretending that last-click attribution captures everything that matters.

Platforms themselves are evolving, and where you invest your measurement effort matters. Newer platforms like Threads present measurement challenges of their own, and understanding what you can and cannot track before committing budget is basic due diligence. The measurement infrastructure should inform the channel decision, not follow it.

Common Measurement Mistakes and How to Avoid Them

Vanity metrics are the most persistent problem. Follower counts, likes, and impressions are easy to report and easy to grow with the right tactics. They are also easy to inflate artificially and have a weak relationship with business outcomes. I have worked with businesses that had grown their social following substantially over several years and had almost nothing to show for it commercially. The audience was real. The content was fine. But there was no clear path from the social activity to anything the business actually cared about.

Conflating reach with impact is a related problem. Reaching a large audience is only valuable if that audience is relevant and if the content is doing useful work. Broad, low-relevance reach is worth less than narrow, high-relevance reach almost every time. When you are evaluating social performance, ask whether the people you are reaching are the people you need to reach, not just how many of them there are.

Over-reliance on single-platform reporting is another trap. Every major social platform has a vested interest in showing that it is driving results. Cross-referencing platform data with independent analytics, CRM data, and revenue reporting is not optional if you want an accurate picture. Third-party social media marketing tools give you a more neutral view of performance across platforms, which is worth the additional complexity.

Finally, measuring too soon is a mistake that kills otherwise sound strategies. Social media, particularly organic social, works on a longer time horizon than most businesses are comfortable with. If you evaluate a social strategy after six weeks and conclude it is not working because you cannot see a revenue impact, you are applying the wrong measurement window. Set expectations for what you will see and when before you launch, not after you have already decided the results are disappointing.

There is more on building a coherent approach to social, from strategy to execution to measurement, across the Social Growth and Content section of The Marketing Juice. If measurement is the question you are trying to answer, channel strategy and content approach are usually the upstream problems that make measurement harder than it needs to be.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions