AI SDR for B2B teams

The AI SDR that starts with a reason to reach out

Most AI SDRs write more emails to the same cold lists. Startories starts one step earlier: it spots a company that just showed a buying signal, finds the person who decides, and writes to them about that moment.

Get started See pricing $1 for 3 days, then $99/month

Updated · 12 min read

What should an AI SDR actually do?

A sales development representative (SDR) has one job: turn strangers who fit your market into qualified conversations. An AI SDR is software that takes over the repetitive parts of that job: researching accounts, finding contacts, writing first emails, following up and sorting replies.

The weak point of most tools is not the writing. It is the list. When every vendor buys the same database and sends the same "quick question" email, buyers learn to ignore it, and inbox providers learn to filter it. Better copy cannot fix a message that reaches someone with no reason to care this week.

Startories treats prospecting as the core of the job. The AI SDR only writes to a company when there is a public, recent reason to do so, and it shows you that reason next to every lead. If you want the background first, read what an AI SDR is and how it works.

AI SDR, AI BDR, AI SDR agent: is there a difference?

Not one that should change your buying decision. In many sales teams, SDRs qualify inbound leads and BDRs (business development representatives) prospect outbound; in others the two titles describe the same job. Vendors borrow both. "Agent" usually signals that the software acts on its own (sending, following up, labeling replies) instead of drafting text for a person to paste. Startories sits on the outbound side: it finds the companies and writes first.

How does the Startories AI SDR work?

  1. Detects buying signals

    It watches Reddit, X, Product Hunt, startup and industry directories, and search results built for your niche. It looks for moments such as a complaint about a competitor, a request for a tool recommendation, a launch, a new hire or a funding round.

  2. Qualifies the company

    Each signal is matched to a real company and checked against your ideal customer profile. The lead gets a score with the reasons written out, so you can see why it was kept and correct the targeting when it is wrong.

  3. Finds and verifies the decision-maker

    It identifies the person who owns the problem (founder, head of growth, CTO, agency owner) and verifies their business email before anything is sent. Unverified addresses are not contacted.

  4. Writes from the context

    The first email refers to the event that triggered the lead, not to a generic compliment about the company. You can approve each email before it goes out, or let a proven funnel run on its own within its limits.

  5. Sends, follows up and sorts replies

    Sequences run from dedicated, warmed-up inboxes on separate domains and stop as soon as someone replies. Every reply is classified (positive, interested, neutral, negative, out of office, unsubscribe) with a suggested answer, so you only spend time on real conversations.

What does one lead look like, from signal to booked call?

Here is a single lead followed through every step. The company, the person and the post are invented for this page; the steps, and what you see at each one, are how Startories works.

Assume you sell usage-based billing software to B2B SaaS companies with 10 to 100 employees in North America, and your ICP excludes consumer apps and agencies.

The signal

On a Tuesday morning, Dana Ruiz, co-founder of a 20-person API company called Fernway, posts on X: "We move to usage-based pricing next quarter and our billing setup cannot meter API calls. What are people using for metered billing?" It is a software recommendation request with a deadline attached: she has already decided to change something and is collecting names.

The qualification

The AI SDR resolves the post to Fernway, checks the company against your ICP and writes down why it kept the lead. In this example the reasons read like this:

  • Company: Fernway, B2B API product, about 20 employees, US. Inside your ICP.
  • Signal: an explicit request for a tool in your category, posted today, with a link to the original post.
  • Timing: a stated deadline (next quarter), so the need will not fade on its own.
  • Contact: the author is a co-founder, so she is likely the person who signs.

The contact

Because the founder asked in public under her own name, she is the right person to write to. The AI SDR finds her business email and verifies it. Had the address failed verification, nobody at Fernway would have been emailed.

The first email and two follow-ups

  • Subject: metered billing before next quarter
  • Email 1 (day 0): "Hi Dana, you wrote that Fernway moves to usage-based pricing next quarter and the current setup cannot meter API calls. That switch is what we build for: metering, invoices and a usage page for your customers, without rebuilding checkout. Most of the work sits in two decisions, how to price overages and what happens to current annual plans. I wrote a one-page checklist for both. Want it?"
  • Email 2 (day 3): "One question decides most migrations like yours: do existing annual customers move to metered pricing at renewal, or right away? The checklist covers both paths."
  • Email 3 (day 8): "Closing the loop. If billing is already handled, ignore this. If not, reply with the word checklist and it is yours, no call needed."

Why these lines work

The first line restates her situation in her own terms, so she knows within a few words why you are writing. The offer is a document, not a demo, which is easier to accept from a stranger. Each follow-up adds one new idea instead of "just bumping this", and the last one makes saying no as easy as saying yes.

The reply

Dana answers on day 3: "Send it. Can it handle prepaid credits too?" The sequence stops. The reply is labeled interested and arrives with a suggested answer: the checklist, a gap for the credits question (only you know your roadmap) and a proposal for a 20-minute call. You fill in the gap and send it. From here, the funnel is judged on the meeting, not on the three emails.

Why do signals beat bigger lists?

Timing is the variable most outbound ignores. A company that posted yesterday that its current tool is too expensive, or that it is hiring its first marketer, is in a different place from a company that simply matches your industry filter. The first one is already looking. The second one has to be convinced that it has a problem.

This is the idea behind signal-based outbound: fewer contacts, chosen for a reason, reached quickly. Our guide to buying signals lists the signals that tend to convert and how to word an email around each of them.

Two ways to run an AI SDR
List-based AI SDRSignal-based AI SDR (Startories)
Who gets contactedEveryone matching firmographic filtersCompanies that showed a recent, public reason to buy
First line of the emailA compliment or a generic hookThe event that triggered the lead, with a link to it
VolumeHigh, to make the numbers workLower, chosen for timing
What you reviewCopy variationsThe reason each company was picked
Main riskSpam complaints and burned domainsFewer leads in a very narrow niche

Reach companies with a reason to buy this week

Startories finds the buying signal, verifies the decision-maker and runs the outreach until they book a call.

Which parts of the SDR job does it take over, and which stay with you?

Write down the job description you would hand a new SDR, then check who does each line. With Startories, most of the hours move to the software. The judgment stays with you.

The SDR job, split between the AI SDR and you
TaskWho does itYour time
Build and refresh the prospect listThe AI SDR, from live signals and niche searchesNone after setup
Check each company against the ICPThe AI SDR, with written reasonsSpot checks, mostly in the first weeks
Find the decision-maker and a working emailThe AI SDR, verified before any sendNone
Write first emails and follow-upsThe AI SDR drafts themApproving drafts until a funnel earns the right to run alone
Run inboxes, warm-up and daily limitsStartoriesNone
Read and sort repliesThe AI SDR labels each one and suggests an answerAnswering the real conversations
Choose the market, the offer and the ICPYou, with the Startories team reviewing early targetingA few hours at setup, then regular reviews
Run discovery calls and closeYou or your account executiveThe meetings themselves

When a human SDR is still the better choice

A good human SDR is still better at live calls, complex accounts and judgment in unusual situations. An AI SDR is better at the volume of research, the speed of reaction and consistency: it reads thousands of posts a day, reacts within hours and never skips a follow-up. If your motion depends on cold calls or LinkedIn steps, a person has to run them: Startories works by email and automates neither.

Most small teams do not have to choose: the AI SDR fills the calendar and a founder or account executive takes the calls. We compare both options, with costs, in AI SDR vs human SDR.

How does it protect your domain and stay compliant?

An AI SDR that damages your domain costs more than it brings. Since 2024, Gmail and Yahoo ask bulk senders to authenticate their mail, keep spam complaints low and offer an easy unsubscribe (Google sender guidelines, Yahoo sender best practices). Startories is set up around those rules.

  • Emails go out from separate sending domains and warmed-up inboxes, never from your main domain.
  • Daily volumes are capped per inbox and sequences stop on reply.
  • Sending pauses automatically when bounces rise, before the problem spreads.
  • Every email includes an opt-out, and unsubscribed or bounced contacts are suppressed across all campaigns.
  • Only verified business emails are contacted, and personal data can be deleted on request.
In the US, commercial email must also follow the CAN-SPAM Act: an accurate sender, an honest subject line, a physical address and a working opt-out (FTC compliance guide).

What do founders worry about before turning on an AI SDR?

These four objections are worth settling before you start. Here are straight answers.

"Prospects will see that a machine wrote it."

People spot generic emails, whoever wrote them. What gives an AI SDR away is the empty compliment and the made-up reason to write. An email that cites a post from two days ago, in plain words, reads like someone did their homework, because the homework was done. In approval mode you can rewrite anything that does not sound like you, and the angle is always the real event, never an invented connection.

"We tried an AI SDR and it emailed the wrong people."

That is a targeting failure, and it is hard to fix when you cannot see why a company was chosen. Here every lead carries its reasons. If you see "kept: offers marketing services" next to an agency you would never sell to, you know which rule to change, and you change it before the next batch instead of after a week of sends.

"Our market is too small."

A narrow market produces fewer public signals, so judge it on the first batch of leads, before anything is sent. If the batch is thin, add a second signal type or niche search prospecting, which finds companies that fit even without a recent event. If it stays thin, your market may be better served by a short list worked by hand, and it is better to learn that in week one.

"I do not have time to review everything."

Review is heaviest in the first two weeks, while you agree on what a good lead and a good email look like. After that, funnels that keep producing leads you agree with can run within their limits. If you want none of the setup work, the done-for-you service handles ICP research, copy, domains and launch, and fully managed acquisition runs the whole program for you.

Who uses an AI SDR like this?

Startories is built for B2B companies with a clear buyer and deals worth at least several hundred dollars. Typical users are founder-led SaaS companies, marketing agencies, software development agencies and consultants. It is not built for consumer (B2C) outreach. What each of them points the AI SDR at differs:

  • Founder-led B2B and AI SaaS: requests for tools in your category and complaints about the product you replace. The founder takes the calls.
  • Marketing and SEO agencies: niche search prospecting for businesses whose websites fit your service, plus companies hiring their first marketer.
  • Dev and AI agencies: fresh launches and funding rounds, which point to teams that need to ship faster than they can hire.
  • Consultants and B2B services: expansion into a new market or segment, which tends to come with a project and a deadline.
If you would rather not run anything yourself, the same engine is available as a done-for-you outbound service: our team handles targeting, copy, sending and replies, and reports on meetings.

What does it cost, and when does it pay for itself?

Plans start at $99 per month, and your first project on Starter starts with a 3-day full-access trial for $1. Starter covers one funnel and about 250 qualified and enriched leads a month; Growth ($499) and Scale ($999) add funnels, sources and volume. See all plans. If you want help, a done-for-you setup costs $1,500 to $2,500 once, and fully managed acquisition starts at $1,999 a month.

To judge the price, work out what one meeting is worth to you. Assume a first-year contract of $6,000 and that one qualified meeting in four becomes a customer. Each meeting is then worth about $1,500 in first-year revenue ($6,000 × 0.25). At that value, Growth breaks even at one meeting every three months, and Starter at one every five. Swap in your own numbers: break-even meetings per month = plan price ÷ (deal value × close rate).

That math leaves out two real costs. One is your time on approvals and calls. The other is the sales cycle: if deals take 90 days to close, revenue from this month's meetings arrives a quarter later, so plan your cash for it. For the cost of the human alternative, see what an SDR really costs.

Lead databases, email finders and sending tools are included, so you do not need separate subscriptions. If you are comparing vendors, our list of the best AI SDR tools and the 11x alternatives page show where Startories fits and where another tool is a better choice.

Frequently asked questions

What is an AI SDR?

An AI SDR is software that does the work of a sales development representative: it researches accounts, finds contacts, writes and sends prospecting emails, follows up and sorts replies, so that people only handle qualified conversations.

How is Startories different from other AI SDR tools?

Startories starts from buying signals instead of a static list. It only reaches out to companies that showed a recent, public reason to buy, shows you that reason for every lead, and references it in the first email.

Will the AI SDR send emails without my approval?

Only if you want it to. You can approve every email and every reply before it goes out, or let a funnel that already works run automatically within its sending limits.

Do I need a lead database or a cold email tool as well?

No. Signal discovery, company matching, decision-maker search, email verification, writing, sending, follow-ups and reply handling are all included in the plans.

Is an AI SDR safe for my domain?

Emails are sent from separate domains and warmed-up inboxes with capped daily volumes. Sequences stop on reply, sending pauses when bounces rise, and every email carries an opt-out.

How many meetings will an AI SDR book?

It depends on your market, your offer and how narrow your ideal customer profile is. Startories reports on meetings and replies rather than sends, so you can judge each funnel on results and stop the ones that do not work.

Sources

Turn fresh buying signals into booked calls

Signals, qualification, verified decision-makers, personalized outreach and reply handling in one engine. Start your first project at $1 for 3 days, then $99/month.