Method + Deck
How an SDR Team Executes Inside an ABM Strategy
My method, applied at several companies: how the GTM team aligns by tier, when the SDR actually enters an account, and what execution looks like from account selection to the first meeting.
Published by Elric Legloire · First presented (in French) on the GTM mon amour podcast · Version française
Navigate with the arrow keys (click the deck first). 15 slides. The full method is written out below.
What ABM actually is
ABM, to me: how the GTM team (Marketing, Sales, SDR, AM, CS) aligns by tier, to maximize our chances of winning new logos and expanding existing customers.
This is about your highest-LTV accounts. The acceptable CAC is higher, and that is what funds the depth. And depth is required: an enterprise B2B purchase involves 5 to 11 decision makers spread across 5 different functions (Gartner B2B Buying Report, 2024).
The tier decides who is involved and which channels you stack
Tier = fit x LTV. Channels are expensive, so you stack them where the LTV justifies it.
- Tier 1, strategic / must-win accounts: AE + SDR + ABM + field marketing + execs + partners. Email, cold calls, LinkedIn, exec-to-exec, direct mail, partner intros, dedicated landing pages, per-account ads, invite-only events.
- Tier 2, high-priority ICP: AE + SDR + ABM + field marketing + PMM. Email, cold calls, LinkedIn, competitor displacement campaigns, webinars, field events, targeted ads, pain-based content.
- Tier 3, scaled intent / retargeting: SDR + ABM + AE. Email sequences, cold calls, LinkedIn touches, retargeting ads, content nurture, webinar invites, intent-triggered campaigns.
One more thing the tier decides: the level of personalization. With AI, personalization is no longer the bottleneck.
And the SDR's scope is wider than new business. On the full bow tie (Winning by Design), the SDR works both sides: new logos and customer expansion.
The problem: mid-market motions on enterprise accounts
Most companies think they run an enterprise motion. In practice they work 500 accounts a quarter with 1 decision maker per account and a 30-day-cycle mindset, on accounts that involve 5 to 11 decision makers and 6 to 18 month cycles.
The data backs the fix: multi-threading lifts win rates by 130% on deals over $50K, and won enterprise deals average 17 contacts (Gong, 1.8M deals analyzed, 2024). One contact, one channel, and zero Marketing / Sales / Post-Sales alignment is what kills these deals, not the length of the sales cycle.
The SDR enters an account the team already prepared
By the first contact, the buyer is already 61 to 69% through their journey (6sense Buyer Experience, 2025). So the sequence matters: leadership selects the account, marketing warms it up, signals accumulate, and then the SDR starts, armed with an account plan and messaging.
The Snowflake model shows the payoff: monthly sales + marketing selection, about 3 weeks of ABM air cover, then the SDR goes in. Result: 36 meetings per 100 accounts under ABM versus 10 without, and deals close 2x faster from SQO to Won (Snowflake ABM team, cohort A/B test, 2024).
Account selection: led by leadership, driven by data
Selection is the biggest lever in the whole system: Tier 1 accounts convert 5x better than lower-tier accounts. It is leadership's job, done with data, and it works as a funnel: a raw TAM of 36,000 accounts filters down to 3,000 actually relevant, then to the top accounts where the team invests (public Castordoc case). Value per account is seats x ACV, not a Tier 1/2/3 label.
Pick the criterion that predicts revenue, not the easiest one to filter on. At Chili Piper, "enterprise" by headcount meant 2,000 employees but sometimes 20 salespeople. The fix: target accounts with 300+ people in GTM.
Map the committee, not a person
A single thread dies with every departure or reorg, and each level of the org hears a different message. So the map has three layers. At the top, executive engagement and sponsorship, owned by the AE and an exec sponsor, with a message about strategic impact, ROI, and vision. In the middle, VPs, where the SDR does 75% of the meeting booking with a message about operational efficiency. Below the power line, directors, managers, and ICs: the SDR collects context and identifies problems to earn the right to go up.
Top-down first. Bottom-up when it stalls. Execs say it themselves: a CRO on LinkedIn wrote that he opens nothing unless a director or VP in his org recommends it, and a Snowflake VP said the best way to get his attention is his team bringing him the topic. One account gives you 100+ entry points.
The account plan: where everything lands
Four blocks: account research (10-K, earnings calls, news, org chart, history with you), contact research (key stakeholders, their roles, what matters to each), your POV (where you can help, why them, why now), and actions (outreach, demo, content, events: who does what, when). The POV is the most underrated step in outreach.
Execution: the channel follows the buyer
Never a single channel. Email, LinkedIn, phone, office visits, personalized sites, gifting, events: the channel follows the buyer, not the rep's preference. Two proof points: dinner invitations during a trade show booked 4 meetings across my top 10 accounts in Europe (May 2026), and 1:1 web pages per account drove 150% more qualified pipeline in the Mutiny x Snowflake case study.
Two real plays from my own accounts. At Castor, one piece of research (a podcast interview with a Gong data leader, run through AI to extract what was relevant to my product) became a 12-second message, served 6 different prospects at the same account, and booked a meeting in 2 hours: my biggest opportunity of the year, at $75K. And a $300 replica Quake rocket launcher sent to Vanta's CRO, a former professional Quake player, got a LinkedIn post and the meeting. On an account worth $200K to $1M in ARR, that gift cost less than the cold email sequence it replaced.
If you remember one thing
Pick the account before the person. The rest is execution: the same method works in the US or in Europe, because it is a system, not a set of tricks.