---
title: "Outreach Benchmarks 2026: The Honest Numbers"
url: https://trysumora.com/blog/outreach-benchmarks-2026
markdown-of: https://trysumora.com/blog/outreach-benchmarks-2026
---

# Outreach Benchmarks 2026: The Honest Numbers

[Home](/)/[Blog](/blog)

Blog

# Outreach benchmarks 2026: the honest numbers

Most outreach benchmark posts inflate numbers. This one publishes what we actually enforce and observe: per-platform send ceilings, funnel-stage rates we track internally, and why industry benchmarks mislead.

By Devansh Ranjan · Published Aug 23, 2026

Most benchmark posts have a product to sell and a funnel they didn't define. This one takes the opposite split: **capacity numbers we enforce and can prove**, and **conversion numbers we measure with a stated method** — clearly labeled as internal.

## The capacity benchmarks (verifiable)

These are the ceilings Sumora enforces per account in production — the same numbers any careful operator should treat as physics:

Platform

Account state

Ceiling

LinkedIn invites

Free, warmed

~20/day, ~100/week

LinkedIn invites

Premium

~35/day

LinkedIn invites

New seat

5/day, ramping over ~12 active days

Instagram DMs

Fresh

~5/day

Instagram DMs

Aged

~30/day (no paid tier lifts it)

Sources and rationale: [LinkedIn limits](/guides/linkedin-connection-request-limits), [Instagram DM limits](/guides/instagram-dm-limits), [warm-up ramp](/guides/linkedin-account-warmup).

## The conversion benchmarks (methodology-defined)

We track a distinct-lead funnel — contacted → reply received → booking proposed → booking confirmed — with A/B arms deciding winners on relative booking-rate lift. Two observations we'll stand behind, with the method published:

-   **Asset openers beat template openers** by roughly 40% relative on reply-to-booking in our campaigns — the measurement is defined in [the lead magnet guide](/guides/personalized-lead-magnets).
-   **Narrow beats broad**: city-level niche ICPs outperform role-only ICPs on reply quality — the targeting that produces them is [geo targeting](/outreach).

## Why industry averages mislead

1.  **Undefined funnels** — a "reply" that includes auto-replies and a "meeting" that includes no-shows are not your numbers.
2.  **ICP contamination** — the average mixes precision-targeted campaigns with sprayed ones.
3.  **Survivorship** — benchmark posts surface tools whose users self-report wins.

The benchmark that matters is your own, defined the same way every week. That's the loop our [AI SDR architecture](/learn/what-is-an-ai-sdr) runs — and the numbers above are the physics it respects ([published capability declaration](/capabilities)).

## Frequently asked questions

What's a good reply rate for LinkedIn outreach?

Honest answer: it depends on niche, targeting precision, and message quality — published 'averages' mix all three. What's universal is the capacity math: ~20 invites/day warmed, so quality per message is the lever, not volume. Reply-rate claims without a defined funnel are marketing.

What daily send limits apply in 2026?

LinkedIn: ~20 connection requests/day warmed free (~100/week), ~35/day Premium, weekly ceilings ramping with account maturity. Instagram: ~5 cold DMs/day fresh ramping to ~30/day aged. Full tables: our LinkedIn and Instagram limits guides.

Why don't you publish a reply-rate benchmark?

Because a defensible one has to define the funnel — contacted → replied → proposed → booked — and disclose the ICP and message mix. Ours is defined in our measurement methodology; industry one-liners generally aren't.

----

_Markdown twin auto-generated from the HTML page at build time. Source of truth: the HTML/_next data._