Skip to content
AI Operations

AI SDRs: What They Can Automate, What Still Needs a Person and the Rules to Follow

An AI SDR can research, draft and sequence outreach, but consent, accuracy and judgement still need people. How to use one without hurting your brand or breaking rules.

Sales development rep reviewing an AI-drafted outreach email on screen beside a printed prospect profile

Field note

Illustrative image, AI-generated. It is not a client, dashboard or result. How we use imagery.

By XenGrowth EditorialPublished Reviewed 10 min read

Share this

Key takeaways

  • An AI SDR is software that researches prospects, drafts outreach and manages follow-ups. It is a tool for a sales team, not a replacement for judgement.
  • Automate research, enrichment, routing and first drafts. Keep a person on positioning, replies to interested prospects and anything that makes a claim.
  • Email, calling and messaging rules apply. CAN-SPAM requires honest headers, a postal address and a working opt-out; calls with AI voices fall under TCPA consent rules.
  • Measure replies that become qualified meetings, not emails sent.

01

What is an AI SDR?

A sales development representative qualifies leads and books meetings. An AI SDR automates parts of that work: finding and enriching prospects, drafting personalised messages, scheduling sequences, logging activity and flagging replies. Some tools send autonomously; most responsible setups route drafts to a person first.

The value is time. Reps spend much of the day on research and admin. Automation can return that time, but only if the targeting and message are right to begin with. Software cannot repair a weak offer, only send it faster.

02

What should be automated and what should not?

Automate tasks that are mechanical and checkable. Keep human control where trust or accuracy is at stake. A wrong claim about a prospect's company or a clumsy reply to an interested buyer costs more than the minutes saved.

Treat the CRM as the source of truth. Log every automated touch, suppress contacts who opted out, and make it easy for a rep to see what was sent and why.

What should be automated and what should not? comparison table
TaskAutomate?Reason
Account research and enrichmentYesMechanical; verify sources
First-draft personalisationYes, with reviewCheck facts before sending
Sequencing and schedulingYesRules-based
Replies to interested prospectsPersonHigh value, nuanced
Pricing, terms and promisesPersonCommitment and liability
Opt-out handlingYes, enforcedLegal requirement

03

What rules apply?

In the US, the CAN-SPAM Act sets requirements for commercial email: no deceptive headers or subject lines, a valid physical postal address and a clear way to opt out that you honour promptly. Other regions, such as the EU and Canada, have their own consent rules that differ from CAN-SPAM, so check where your prospects are with a qualified adviser.

For phone outreach, the FCC has confirmed that AI-generated voices in calls count as artificial voices under the TCPA, which requires prior consent for many such calls. Do not use AI voice dialling at scale without legal advice on consent.

  • Honest sender identity and subject lines
  • Physical address and working unsubscribe
  • Suppression list shared across tools
  • Consent records for calls and texts
  • Regional rules checked before launch

04

How do you judge whether it works?

Track reply rate, positive reply rate, meetings booked, meetings held and opportunities created, by segment and by message. Compare against your human baseline for the same accounts. Sending volume is not a result.

Watch brand signals too: spam complaints, bounce rate and unsubscribes. If they rise, you are trading long-term sender reputation for short-term activity. Slow down and fix targeting.

Was this useful? Send it to someone who owns this problem.

Primary sources and further reading

Use the source material to validate details against your own context and current platform configuration.

This field note follows the XenGrowth editorial policy: primary sources where available, visible limitations, material review dates, and no invented first-hand experience.

Stay with the problem

Explore AI & automation