RTB vs. Direct Call Buying: Which Model Wins?

Two ways to buy calls

If you buy inbound phone calls at scale, you have two procurement models: real-time bidding (RTB) through an exchange, or direct deals negotiated with individual publishers.

Both work. Both have been around for years. But they produce very different economics, and most buyers default to one model without fully understanding the trade-offs. This article breaks down the comparison across five dimensions that matter most to demand-side call buyers: pricing transparency, quality control, scale, optimization speed, and attribution.

Pricing transparency

RTB: Second-price auctions

In an RTB exchange, every call goes through an auction. Multiple buyers submit bids, and the winner pays the second-highest bid plus a small increment -- not their maximum bid. This is the same mechanism that powered Google AdWords for over a decade, and it exists for a reason: it incentivizes truthful bidding.

If a Medicare call is worth $45 to you and $32 to the next buyer, you bid $45 and pay $32.01. You don't overpay. The market sets the price, and you can see exactly what the clearing price was for every call.

RTB exchanges also publish historical clearing prices by geo, vertical, source, and time of day. You can model expected costs before you launch a campaign. There are no surprises.

Direct deals: Negotiated rates

Direct publisher deals use fixed pricing -- a negotiated CPL (cost per lead) or per-call rate that stays constant for the duration of the contract. Common structures include flat per-call pricing (e.g., $28/call regardless of quality), tiered pricing based on call duration or disposition, and monthly minimums with volume discounts.

The advantage is predictability. You know exactly what you'll pay. The disadvantage is opacity. You have no idea whether you're paying market rate or 40% above it. The publisher knows their costs and margins. You don't.

Direct deals also create adverse selection risk. If you negotiate a $28 flat rate and the publisher's average cost to generate a call drops to $12, they have no incentive to pass savings along. Conversely, if their costs rise, they'll push to renegotiate or start sending lower-quality traffic.

Verdict: RTB wins on transparency. Second-price auctions ensure you pay fair market value on every call, with full visibility into clearing prices.

Quality control

RTB: Propensity scoring and pre-auction filtering

Modern RTB exchanges evaluate caller quality before the auction runs. Propensity scoring models analyze caller metadata -- geo, traffic source, time of day, historical conversion patterns for similar callers -- and assign a quality score to each call opportunity.

As a buyer, you use these scores to make informed bid decisions:

• Score 85+: High-intent caller from a proven geo/source combo. Bid aggressively.
• Score 50-84: Mixed signals. Bid at base rate.
• Score below 50: Low-intent or high-fraud-risk. Skip automatically.

This scoring happens in milliseconds, before you commit budget. You can also set source-level filters, geo restrictions, and daypart rules at the campaign level. Every call you buy passes through multiple quality gates.

The scoring models also improve over time. As you feed conversion data back into the system, the models learn which caller profiles convert specifically for your business, not just the market average.

Direct deals: Manual QA and trust

With direct publishers, quality control is reactive. You buy calls, listen to recordings, flag bad ones, and request credits. This cycle typically takes 3-5 business days per dispute, and there's always friction over what counts as a "bad" call.

Some publishers offer basic quality filters -- minimum call duration, caller ZIP code restrictions, IVR pre-screening. But these are blunt instruments compared to ML-based propensity scoring. A 90-second call from the right ZIP code can still be completely unqualified.

The core problem is information asymmetry. Publishers have data about their traffic sources that they don't share with buyers. When quality drops, you find out after the money is spent.

Verdict: RTB wins on quality control. Pre-auction scoring and automated filtering prevent bad spend rather than just recouping it after the fact.

Scale

RTB: Exchange volume

RTB exchanges aggregate supply from hundreds of publishers, traffic sources, and networks. A single exchange might process 50,000 to 500,000 call opportunities per day across all verticals. You access this entire pool through one integration.

Need to scale from 200 calls/day to 2,000? Raise your bids and budgets. The exchange handles sourcing. You don't need to find, vet, and contract with new publishers individually.

Geographic expansion is equally simple. If you're buying Medicare calls in Florida and want to add Texas, you create a new campaign targeting TX ZIP codes. The exchange already has Texas supply from dozens of publishers.

The trade-off is that exchange volume fluctuates. Seasonal verticals like Medicare AEP see massive supply spikes in Q4 and drops in Q1. You're competing with other buyers for the same pool, so costs rise when demand is high.

Direct deals: Publisher caps

Direct publisher deals have hard volume caps. A publisher generating 100 Medicare calls per day in Florida can't suddenly deliver 500 because you increased your budget. Publisher volume is bounded by their traffic generation capacity -- their ad spend, their SEO rankings, their call center infrastructure.

Scaling through direct deals means adding more publishers. Each new publisher requires discovery, vetting, test campaigns, contract negotiation, and ongoing management. Adding 10 new publishers might take 2-3 months and significant account management overhead.

The upside of direct deals is supply exclusivity. If you have an exclusive agreement with a high-quality publisher, competitors can't bid on those calls. This matters in competitive verticals where the best publishers are capacity-constrained.

Verdict: RTB wins for scale and speed-to-volume. Direct deals win for exclusive access to specific high-quality sources, but can't match RTB for rapid scaling.

Optimization speed

RTB: 15-minute AI cycles

RTB platforms with AI optimization engines can test and apply bid changes in near real-time. A typical cycle looks like this:

1. Minute 0: System identifies that Florida calls are converting 2.4x above average.
2. Minute 5: System proposes a +25% geo bid modifier for FL.
3. Minute 10: Experiment runs on a deterministic traffic split.
4. Minute 15: Results evaluated. If CPA improves, modifier is applied automatically.

This means your campaigns can react to intraday shifts -- a surge in high-quality callers from a particular source, a drop in conversion rates during a specific daypart, a new competitor entering the auction. Adjustments happen in minutes, not days.

The compounding effect is significant. Over a week, a platform running 15-minute optimization cycles tests and applies hundreds of micro-adjustments. Each one is small, but the cumulative impact on CPA can be 10-20%.

Direct deals: Weekly reviews

With direct publishers, optimization is a manual process. The typical cadence: pull a weekly report, analyze conversion rates by publisher, identify underperformers, email the account manager requesting adjustments or credits, wait for a response, implement changes.

This cycle takes 5-10 business days. During that time, you're still paying the same rate for traffic that may have degraded. If a publisher's traffic source changes mid-week -- they shift ad spend from Google to a lower-quality network -- you won't catch it until the next review cycle.

Even with dedicated account management, the feedback loop is fundamentally slower. Humans reviewing data weekly can't compete with algorithms testing continuously.

Verdict: RTB wins decisively. Automated optimization at 15-minute intervals produces compounding improvements that manual weekly reviews cannot replicate.

Attribution

RTB: Ring pools and full-funnel tracking

RTB platforms integrate with ring pool technology to capture click-level attribution. Here's the flow:

1. A visitor clicks a Google Ad and lands on your page. The ring pool assigns a unique tracking number and captures the gclid.
2. The visitor calls the tracking number. The call enters the RTB auction.
3. You win the auction. The call is routed to your agents.
4. The call converts to a sale.
5. You report the conversion through the platform.
6. The platform fires a postback to Google Ads with the original gclid and the revenue value.

Google now knows that this specific keyword, in this specific ad group, drove a sale worth $X. Smart Bidding uses this signal to optimize future ad delivery. The same flow works for Meta CAPI, Bing UET, and TikTok Events API.

This creates a true closed loop: ad click, call, sale, and postback all connected through a single attribution chain.

Direct deals: Static numbers and partial data

Direct publisher deals typically use static tracking numbers -- one number per publisher, sometimes one per campaign. This tells you which publisher generated the call, but nothing about the upstream traffic source.

Did the call come from a Google search ad, an organic listing, a Facebook post, or a bought lead list? With a static number, you don't know. You can't close the loop back to the ad platform because you don't have the click ID.

Some sophisticated publishers will share gclid data, but this requires custom integration work and trust that the data is accurate. Most publishers treat their traffic source data as proprietary.

Verdict: RTB wins on attribution. Ring pool integration with click ID capture enables full closed-loop reporting that direct deals cannot match without significant custom engineering.

When direct deals still make sense

Despite RTB's advantages across most dimensions, direct deals have legitimate use cases:

Exclusive high-volume sources. If a publisher generates 1,000+ calls/day in your vertical with proven 40%+ conversion rates, an exclusive direct deal protects that supply from competition. The premium you pay is insurance against losing access.

Regulated verticals. In verticals like healthcare or financial services where compliance requirements are strict, direct relationships with vetted publishers give you more control over the consumer experience and disclosure requirements.

Test-before-you-bid. Some buyers use direct deals to evaluate new verticals or geos before committing to RTB campaigns. Lower volume, fixed pricing, and a single-publisher relationship simplify initial testing.

Relationship leverage. Long-standing publisher relationships can unlock benefits that exchanges can't provide -- early access to new traffic sources, custom IVR flows, dedicated call center capacity during peak periods.

The bottom line

For most demand-side call buyers optimizing for scale and efficiency, RTB is the superior model. Second-price auctions ensure fair pricing. Pre-auction scoring prevents bad spend. Exchange volume enables rapid scaling. Automated optimization compounds daily. And ring pool integration closes the attribution loop.

Direct deals remain valuable for exclusive supply relationships and regulated verticals, but they shouldn't be your primary procurement channel if you're buying more than a few thousand calls per month.

The optimal strategy for large buyers: run 70-80% of volume through RTB for efficiency and scale, and maintain direct relationships with your top 3-5 publishers for exclusive, high-converting supply.