Batch Instead of Drip Sourcing: Outbound Cadence in Practice
Why lead sourcing in practice happens in bursts rather than evenly: real sourcing data, pros and cons of batch vs. drip, and a decision model.
The myth of the steady lead supply
In theory, a healthy outbound machine looks like this: every day, a constant number of new leads flows in at the top of the pipeline, and every day, contacted accounts leave at the bottom. A calm, predictable stream. Practice looks different — not because teams lack discipline, but because the economics of sourcing force a different cadence.
A look at real sourcing data from a production multi-campaign setup shows the pattern: new leads appear in bursts. Individual days bring 166, 79, or 58 new accounts, with days of few or no new additions in between. This isn’t an outlier, it’s the normal state — and it has good reasons.
Why sourcing naturally tends toward batches
Setup costs are incurred per run, not per lead. A sourcing run consists of segment definition, source selection, filter building, qualification, and import checks. Whether 15 or 150 leads come out at the end changes this setup effort very little. Once a clean segment has been defined — say, “owner-managed machine builders, 20 to 200 employees, DACH” — it makes sense to pull the entire cohort at once, not five companies a day. How to set up such a segment search is shown in Addresses for cold outreach.
Qualification works better in blocks. Anyone looking at 150 candidates side by side spots patterns: duplicate domains, wrong size classes, segment outliers. The cohort can be validated as a whole — including email verification before import, the most important protection for your own sender infrastructure.
Campaigns think in cohorts. A batch sourced as a unit can be evaluated as a unit: same source, same segment, same angle, one result. With drip, cohorts blur, and after four weeks nobody remembers which sourcing decision produced which reply rate.
Batch vs. drip side by side
| Criterion | Batch Sourcing | Drip Sourcing |
|---|---|---|
| Setup cost per lead | Low (spread across a large cohort) | High (setup repeats) |
| Cohort evaluation | Clean (one segment, one result) | Difficult (mixing) |
| Data freshness | Ages if batch too large | Maximally fresh |
| Signal capability | Unsuited to triggers | Ideal for triggers |
| Predictability | High (reservoir visible) | Depends on daily yield |
| Operating rhythm | Focus sessions | Daily routine |
The table shows: there’s no overall winner, but a clear division of labor. Batch wins wherever a defined ICP segment is being worked through methodically. Drip wins as soon as timing determines relevance — that is, everything described in signal-based outbound: a buying signal from yesterday is worth more than a perfect ICP match from six weeks ago.
The decoupling: source in bursts, send evenly
The most common objection to batch sourcing concerns deliverability: doesn’t a 500-lead import burn the domains? No — if sourcing cadence and sending cadence are cleanly separated. The batch fills only the reservoir. The sequence engine then pulls as many first contacts from it daily as the inbox pool can cleanly handle, distributing them across senders and weekdays. Bursty input, even output.
This decoupling is the actual core of the model. It allows combining the advantages of both worlds: the efficiency and evaluability of the batch when building lists, the continuity of the drip when sending — including the conservative per-inbox limits that are mandatory anyway under email deliverability.
A decision model for your own cadence
- ICP segments → batch. Defined segment, source the entire cohort, validate, import. Size the batch to two-to-four-weeks of processing capacity.
- Signals → drip. Skim trigger sources (engagement, visits, events) daily and push them ahead of the batch reservoir with priority — signals expire, segments don’t.
- Sending → always even. Regardless of the sourcing rhythm, sending stays throttled and distributed.
- Evaluation → per cohort. Every batch gets its own evaluation by reply rate and qualified replies, so the next sourcing decision is data-driven.
Anyone trying to run this rhythm manually usually underestimates the coordination overhead between sourcing, validation, import, and send-throttling. In GTM Goat, CegTec’s outbound system, exactly this decoupling is built in: sourcing runs fill segments in bursts, the campaign engine works through them evenly with throttling, and every cohort stays evaluable as a unit. What that feels like can be observed in the four-week free trial — including your own first sourcing batch.
Common questions
What's the difference between batch and drip sourcing?
Batch sourcing builds lead lists in bursts: on one day, 100 or more new accounts are researched, qualified, and imported, followed by a phase in which the pipeline works through that cohort. Drip sourcing, by contrast, feeds the pipeline a small, constant amount of new leads every day. Both models fill the same pipeline, but they differ sharply in effort, data quality, and controllability.
Why does batch sourcing dominate in practice?
Because sourcing effort has setup costs: defining the segment, choosing the source, building filters, checking qualification. These costs are incurred per sourcing run, not per lead. A run with 150 leads costs barely more setup than one with 15. This shows up clearly in real workspace data: new leads appear in spikes of 166, 79, or 58 on individual days, with quiet stretches in between.
Does deliverability suffer when leads are imported in batches?
No — as long as import and sending are decoupled. The batch only fills the reservoir; the sequence engine still distributes sending evenly across days and inboxes. It becomes critical only if a batch import is sent immediately without throttling. More important than cadence is validation: every batch should be checked for bounce risk before import.
When is drip sourcing the better choice?
For signal-based outbound. Anyone reacting to triggers — job changes, funding, website visits, LinkedIn engagement — can't batch, because signals arrive daily and age quickly. Here, a daily drip is the only sensible cadence. In practice, most mature setups run a hybrid model: batch for ICP segments, drip for signals.
How large should a sourcing batch be?
As large as downstream capacity can work through in two to four weeks. Anyone running 30 send-ready inboxes at roughly 15 cold emails per day processes about 450 first contacts daily — a batch of 500 to 2,000 qualified leads fits that. Batches far beyond processing capacity age in the database: contacts change roles, signals expire.