Stackmedia is a small digital agency based in Abuja, Nigeria. Its owner, Tunde, built the business over four years running Meta and Google Ads for local businesses – dental clinics, law firms, HVAC contractors, a few retail shops. The kind of clients who need leads, not impressions. The kind where a bad month gets noticed immediately.
For most of that time, the model worked. Facebook delivered affordable clicks. Google caught people already searching. Results were predictable enough that clients stayed, referred others, and kept paying.
Then it stopped working as reliably as it once had.
CPCs on Google climbed. Meta's algorithm changes arrived faster than Tunde could adapt to them. The budgets his clients were spending were producing fewer leads for the same money, and the clients were asking why. Tunde did not have a satisfying answer. More budget was not a strategy. Blaming the platform was not something he could say out loud.
What he needed was a way to diversify. Not replace Google and Meta – that was never the plan – but add something that could take some of the pressure off. A channel he actually understood and could explain to a dental clinic owner who wanted to know where their money was going.
That search led him to programmatic advertising, and eventually to Epom DSP.
When Meta and Google Stop Working the Way They Used To
The pattern Tunde noticed was not unique to him. Across his client base, the economics of performance advertising were shifting. Auction competition was increasing. More advertisers chasing the same local audiences meant higher CPCs. Platforms that once felt transparent were now offering less and less visibility into why a campaign was or was not performing.
For a small agency managing mostly SMB clients on tight budgets, the margin for error was thin. When a dental clinic's lead volume dropped, the conversation with the client happened fast. There was no buffer, no brand equity to hide behind, no monthly retainer that insulated the agency from the results.
"The platforms kept changing, and we kept reacting. That is not a strategy. I needed something I could add that wasn't just more of the same."
The frustration was not just about performance. It was about dependency. Stackmedia's entire revenue sat on two platforms that could suspend an account, change an algorithm, or raise prices with no notice. That felt increasingly fragile as the agency grew.
Programmatic had been on Tunde's radar for a while. He knew it existed. He assumed it was for bigger agencies with specialist teams and serious budgets. Then he started looking more carefully.
Why Programmatic Felt Too Big – Until It Didn't
The first time Tunde looked into DSP platforms seriously, he nearly talked himself out of it. Enterprise DSPs came with pricing structures that assumed a much larger operation. Some required minimum monthly spends that exceeded what several of his clients spent combined. Others expected the agency to have a dedicated ad ops person – someone who would own the platform full-time.
Stackmedia had Tunde, one junior media buyer, and a part-time analyst. That was the team.
"Every DSP I looked at seemed built for someone with three times my team and ten times my budget. I started thinking maybe programmatic just wasn't for agencies like mine."
What changed his mind was Epom's entry point. The pricing was structured for agencies still testing the channel – no long-term contract demanding commitment before confidence existed, and no minimum client budget that would have disqualified his entire book of business.
It was a platform built to let an agency like Stackmedia start small, prove the value to one or two clients, and scale only once the results justified it. That mattered more than any feature list.
| Option Considered | The Problem |
|---|---|
| Enterprise DSPs (e.g., DV360) | High minimums, complex setup, required dedicated ad ops staff |
| Other mid-market DSPs | Opaque pricing, hidden tech fees, limited transparency on placements |
| Staying on Meta + Google only | Rising CPCs, algorithm dependency, no answer for clients on declining ROI |
| Epom DSP | Low entry cost, self-serve, transparent reporting, human support included |
Getting Started Without an Ad Ops Team
Tunde's first campaign on Epom was for a private medical clinic in Abuja, running a seasonal promotion. The client had a modest budget and a specific goal: appointments booked through the clinic's landing page. Not brand awareness. Not impressions. Appointments.
The setup took longer than it would today – Tunde was learning the platform at the same time as running the campaign. But the interface was navigable without specialist knowledge, and when he got stuck, the Epom support team responded with real answers rather than redirecting him to a help article.
"I had a specific question about bid settings that I couldn't find in the documentation. I messaged support and had an answer within a few hours."
The campaign ran for three weeks. The cost per lead came in 38% lower than the clinic's Meta campaigns were delivering at the same time. That was not what Tunde had promised the client – he had been careful not to overpromise – but it was a result he could show, compare, and build a conversation around.
That first result did two things. It gave Tunde confidence that the channel worked for SMB clients in Nigeria. And it gave him a talking point in every new business conversation that followed.
What Transparency Actually Means When a Client Is Asking Questions
One of the sharpest differences Tunde noticed between Epom and the platforms he had been using was how much he could see. On Meta, optimizing a campaign often meant trusting an algorithm that did not explain itself. On Google, the level of placement visibility depended on how much you pushed for it.
On Epom, he could see exactly where ads were running. Which domains. Which formats. Which placements were spending the budget and what they were returning. That granularity changed what client conversations looked like.
"When a client asks where their money is going, I can now actually show them. That sounds simple but it is a completely different conversation from saying 'Meta's algorithm decided.'"
For local SMB clients, especially in professional services like legal and medical, where brand safety matters, knowing where ads appear is not just a reporting preference. It is a trust requirement. A law firm does not want its banner sitting next to content it cannot stand behind.
The ability to blacklist domains, monitor placement quality, and show clients a clean report of where their budget actually went became one of Stackmedia's selling points for programmatic. Not the technology. The honesty made it possible.
| Factor | Meta / Google Only | With Epom DSP Added |
|---|---|---|
| Placement visibility | Limited; algorithm-controlled | Full domain and source-level reporting |
| Client cost per lead | Increasing month over month | ~30% lower on first campaigns tested |
| Platform dependency risk | 100% reliant on 2 platforms | Diversified across open web inventory |
| Agency differentiation | Same tools as every other local agency | Programmatic as a distinct service offering |
| Support when stuck | Ticket system, slow turnaround | Direct team response within hours |
Running Programmatic Alongside Meta – Not Instead of It
One of the mistakes Tunde was careful to avoid was positioning programmatic as a replacement. His clients trusted Google and Meta. Some had been running on those platforms for years. Suggesting they abandon them entirely would have created resistance and risk that the early programmatic results did not yet justify.
Instead, the pitch was simpler: add a channel, diversify the spend, reduce the exposure to any single platform's algorithm or policy change. For clients seeing rising CPCs on Google, adding programmatic display meant some of the budget could reach audiences at a lower cost per impression – particularly on mobile, where Nigerian users spend the majority of their time online.
"I tell clients: this is not about replacing what works. It is about not having all your eggs in one basket. Most of them understand that immediately."
That framing also made programmatic easier to sell to cautious founder-led businesses. A dental clinic owner who had never heard of a DSP was not going to sign off on moving their entire ad budget to an unfamiliar platform. But allocating 20 or 30 percent of monthly spend to test a new channel – with clear reporting and a low entry cost – was a conversation they could have.
Three clients agreed to that structure in the first quarter. All three saw lower cost per lead than their Google or Meta benchmarks. One expanded their programmatic budget within six weeks of launch.
What Comes Next for Stackmedia
Tunde is not planning to move fast. The agency has a reputation built on reliable results for local clients, and he is not willing to risk that by scaling before the process is solid. Programmatic is being introduced carefully – one vertical at a time, one client conversation at a time.
But the direction is clear. Meta and Google will remain part of the mix. They are too established in Nigeria's SMB market to ignore. What is changing is that Stackmedia no longer depends on them entirely.
"I used to dread the months when Meta performance dropped, and I had nothing else to offer. Now I have a conversation instead of an apology."
The agency is now evaluating programmatic for two more client verticals – home construction services and a local education centre running enrollment campaigns. Both have the kind of audience targeting requirements – geography, device, time of day, contextual placement – where programmatic gives more control than Meta's audience tools currently allow.
Longer term, Tunde is considering moving to Epom's white-label capabilities as client volume grows. The ability to present a branded platform to clients would strengthen the agency's positioning and make programmatic feel less like a third-party tool and more like a Stackmedia product.
For now, the goal is simpler: keep building proof. Every campaign that delivers a lower cost per lead than Meta is a data point. Enough data points become a product.
Stackmedia did not set out to become a programmatic agency. It set out to stop losing clients to a problem it could not solve with the tools it already had.
Epom DSP gave Tunde a way in – low enough cost to test without risk, transparent enough to explain to a dental clinic owner, and supported well enough that a two-person team could run it without hiring a specialist. That combination is not common in a market where most DSP options assume a much larger operation.
The Nigerian SMB market is not waiting for programmatic to arrive. The agencies that figure it out first will have a durable advantage over the ones that stay inside the walled gardens until the economics force them out.