How to Turn ABM Into a Strategy That Works for Marketing and Sales

ABM is a strategy for putting the right amount of time, effort and budget into the accounts with the most potential return, and it only works when marketing and sales run it as one motion. It is not a replacement for consistent calling, outbound and cold prospecting. Most ABM programmes stall because they are a marketing list that sales never adopted, or a sales target list that marketing never supported.

This guide turns one conversation into nine practical steps. It draws on our episode with Jenna Chambers, VP of Marketing, on why ABM approaches fall down for sales development teams, plus what we see when we help teams put ABM into practice. It is written for the marketer, SDR leader or founder who has to make ABM work with the team they already have.

By the end you will have agreed accounts, clear roles, a signals stack you trust, a feedback loop, and a way to prove it is working.

The Stakki View: ABM Allocates Time, It Does Not Replace Prospecting

Jenna's conversation sits on top of a view we hold strongly at Stakki, based on the 150+ sales stacks we have worked on, seen and talked through with guests on the series.

• ABM is a resource allocation strategy. It answers one question: how much time, effort and budget should each account get, based on the return it could deliver later?
• The goal is still focused prospecting. Tiering accounts exists to direct rep time into the right conversations, not to add a layer of activity on top.
• It all falls apart without a sales process that enables consistent conversations. If reps do not have a clear target list, protected calling time and a cadence they actually follow, no account strategy will fix it.
• ABM is not a replacement for consistent calling, outbound and cold prospecting. It tells SDRs and AEs where to aim. The calling, outreach and prospecting still have to happen every week.

Read the nine steps below with that in mind. Each one either decides where time goes, or protects the process that turns that time into conversations.

Step 1: Agree what ABM is for, and what it is not

Start with a shared decision, not a marketing announcement. Jenna's first point is that ABM should come out of a consultative process across marketing, sales, SDRs and customer success, and should never simply be dropped on the SDR leader.

The tension is real. Many SDR teams are built on volume: more reps, more calls, more at-bats. ABM is close to the opposite, with fewer accounts and more depth, so reps worry about hitting target on fewer attempts. Name that worry early.

The model Jenna recommends is blended ABM. You keep wider coverage of your ideal customer profile (ICP) and run ABM on a smaller set of accounts on top of it, so nobody is asked to abandon the day job.

Our own view is simpler than most ABM frameworks. ABM is a way to decide where your time, effort and budget go, based on the return an account could deliver later. It is not a separate motion that replaces consistent calling, outbound and cold prospecting.

Do this: run one 60-minute session with marketing, sales, SDR and CS leads, and leave with a one-page charter that answers four questions.

1. What is ABM meant to deliver: new pipeline, expansion in existing customers, or both?
2. Are we running inbound plays, outbound plays, or both?
3. Which part of our ICP is ABM, and which part stays on regular coverage?
4. What role does each team play, in one sentence each?

Add one more line to the charter: what does not change. Daily calling blocks, outbound cadences and cold prospecting into the wider ICP all carry on. ABM changes where reps aim and how much effort an account earns, not whether they pick up the phone.

Step 2: Choose and tier your accounts together

Not every ABM account deserves the same treatment, so split your list into three tiers before anyone writes a message. Jenna is clear that an ABM account is not always created equal.

The classic failure is going too wide, too fast. Jenna describes teams who tried ABM on a list of 2,000 accounts for three months, saw nothing happen, and concluded ABM does not work.

Tier by potential return, not by logo size. Score accounts on what they could be worth later, how well they fit, and how realistic they are to win, then give the most time, personalisation and budget to the highest scores. That is all ABM really is: a way to spend limited effort where it is most likely to pay back. The wider list still gets worked every week.

Do this: rank your ICP by fit and revenue potential, agree the tier cut-offs with sales and CS, and keep the one-to-one tier small enough that an AE can genuinely know every account in it. At the same time, ring-fence roughly 5% of eligible accounts as a control group (Step 8), so you can prove the programme works.

Step 3: Write the rules of engagement

Write down who does what on each account before outreach starts. Jenna calls these the rules of engagement, and they are the difference between a team and a group of people emailing the same prospect.

In particular, intent alone does not make an account warm. An account can show a clear pain point and still not know who you are, which makes your call a cold call. Marketing needs to be raising awareness in the account first, so the SDR lands softly.

Do this: put these answers on one page and share it with every team.

• Who makes first contact on each tier, and with which personas?
• If a tiered account comes in through inbound, does it go straight to the AE, or does an SDR qualify it?
• Who can contact an account that already has an open opportunity?
• What air cover does marketing run before and during outreach?
• What does customer success do for ABM accounts that are existing customers?
• Who owns each account's plan, and where is it kept?

Revisit the page after the first month. The rules will change once real accounts start to move.

Step 4: Split the work between AEs and SDRs

Give AEs an 80/20 split and point SDRs at the gaps the AEs cannot cover. Jenna's guide for AEs is about 80% of their time on one-to-one and one-to-few accounts, and about 20% on one-to-many.

For SDRs, the highest-value job on the top tiers is to open up the buyer group. If an AE is working with five contacts in a target account, the SDR keeps opening doors beyond them. Another option is to point the SDR at a new region or business unit in the same account, where the AE's existing relationship works as an internal case study: "we are working with your peers."

On the wider list, SDRs still run outbound, but it is no longer a pure cold call, because marketing is already engaging those accounts. As Jenna puts it, "the excellence of SDRs is to be able to bring opportunities to the table that the business otherwise wouldn't be able to bring."

A word of caution from us: ABM tells reps where to aim, but it does not do the calling. It only works if your sales process already produces consistent conversations with your target market, with a clean target list, protected call blocks, a cadence reps actually follow and one place to see activity. If that is shaky, fix it first. Tiering an inconsistent process just gives you a tidier list of accounts nobody is speaking to.

Two traps to coach against:

• Over-dialling signal accounts. If reps only chase accounts showing signals, they can make dozens of calls into one account and neglect the rest of the list. Keep consistent work going across the whole list.
• Treating research as discovery. Research is a starting point. You still need to talk to the person to learn how they see the problem.

Do this: agree the split per role, then put signals and account context on the SDR's working list so everything is visible while the phone is ringing.

Step 5: Build a signals stack you can trust

Treat signals as plural: act on a mix of credible sources, never on one score you cannot explain. Jenna's warning is blunt: "there's a lot of snake oil out there." Her team's name for the result is the data mirage, where the dashboard looks full of signals and the pipeline quietly disappears.

The shift from "intent" to "signals" matters because intent came to mean third-party intent. Signals can come from many places, and the value is in combining them.

Two reasons to be careful with third-party data. A black-box score that hit a threshold, with no explanation of why, should not drive spend. And if you and your competitors buy the same single-source intent, you will all call the minute an account surges.

Do this: audit every signal you use, note where it comes from, and only escalate an account when two or more independent sources agree. Then route the result to where reps already work, such as the CRM and a team chat channel, so the signal arrives with the account.

Tools we recommend for signals. These are Stakki partners, so you should know we have a relationship with them. We only suggest them where they fit the job.

• ClearCue or PhantomBuster for buying signals. Use them to watch for triggers such as hiring, job changes and LinkedIn activity at your target accounts, and feed the results onto your account lists.
• Buyer intent in HubSpot. If HubSpot is already your sales CRM, switch on its buyer intent first so signals land on the account record where reps already work.
• Lead Forensics for website visits. It shows which companies are visiting your site, which covers your first-party signal and helps you spot target accounts that were not visiting before.
• Hotline Ring to close the speed-to-lead loop. A signal is only worth acting on quickly. Hotline Ring rings your team the moment a hot lead comes in, so the account hears from you while it is still looking.

Our rule on tools: start with the CRM and your sales process, find the bottleneck, and only add a tool that solves a clear problem. These are tools we recommend regularly and use ourselves, built around mid-market, cost-conscious teams, and other tools may suit enterprise or early-stage teams better.

Step 6: Personalise at persona-plus-account level

Personalise by combining what the account cares about with what the person cares about in their role, and use AI to make that fast. Jenna's golden rule is to ask what the account cares about, then what the individual cares about, and marry the two. People forget they are writing to an individual at work.

That used to mean deep research on a handful of accounts. Now you can train an AI agent to research any account in minutes, pointing it at:

• company websites, press releases and annual reports
• social posts, including what employees are sharing
• recent job ads for the roles you target, which Jenna calls her favourite tip, because they tend to show what the company is struggling with, since that is what it is hiring to solve

Depth should follow the tier. One-to-one and one-to-few get the most tailoring, one-to-many gets lighter touches. In Jenna's words, there is "kind of no excuse for just like flat generic outreach these days."

Personalisation must never block the call. If reps feel they have to research everything before dialling, the programme slows to a crawl. Put the research where the rep can read it in seconds.

Do this: build one short persona card per role you target, covering what they care about, the problems they hit and how your positioning connects to them. Then test your channel mix, for example comparing calls against email on a defined set of accounts for a few weeks, and keep email deliverability healthy so the messages actually arrive. Finally, still run discovery on every call, because research will not tell you how the person in front of you interprets the same facts.

Step 7: Close the feedback loop

Insight has to flow back from sales to marketing as reliably as leads flow forward. Marketing hands over accounts and signals; the call is where you learn what lands. If SDRs keep those conversations to themselves, marketing never finds out that a glossy new campaign is not landing.

Call recording makes this practical. Jenna advocates recording every call and setting up trackers for the messages you want to hear, so the picture comes from real conversations and not from the marketing team's own assumptions.

She also describes a sharper test. When her team launched new positioning, they scored calls against it, and she plans to track how many of the highest-scoring calls progress to the next stage. If top-scoring calls move forward, the messaging is working. If reps nail the message and deals still stall, the messaging itself needs another look. And if the message is not coming across as intended, that is on marketing for not enabling the team properly.

Do this:

1. Record calls and set trackers for your core positioning, key pain points and competitors.
2. Push tagged calls automatically into the CRM and a team chat channel, so nobody has to log into another tool to find them.
3. Hold a 30-minute weekly review where marketing and sales listen to a handful of calls together and note what to change.
4. Use AI role-play practice to coach reps on the messages that are not landing.

Sync notes and calls back to marketing with Claude MCPs. An MCP lets Claude read and work with a tool in plain English. Connect it to your dialers and conversation intelligence tools, such as Fathom, CloudTalk and Rocketphone, and it can pull call summaries and sales notes back to marketing without anyone logging into another platform. Fathom and CloudTalk both have official MCP servers (Fathom, CloudTalk).

Two asks to try:

• "Pull this week's calls with our tier 1 accounts and summarise what prospects said about our new positioning."
• "List the objections raised on target-account calls and write a short weekly summary for marketing."

Start with read-only access, and review the first few outputs before you let it update anything.

Step 8: Measure leading indicators and hold back a control group

Track leading indicators from the first month and keep a control group so the results are not a matter of opinion. ABM is slower and more expensive than volume outreach, and senior leaders will want evidence before the revenue shows up.

We would add two measures that ABM programmes often miss. The first is reach rate: of 1,000 prospects on your list, how many do you eventually reach and speak to? We rate this above connect rate (dials per conversation), which most teams confuse with it. The second is activity consistency: check that calls and outreach per rep have not dropped since ABM started. If they have, ABM has quietly replaced prospecting instead of focusing it.

Then add the comparison that makes it credible. Jenna's advice is to keep a small portion back for an A/B test, so you have a baseline: run ABM plays on about 95% of the ring-fenced accounts and leave about 5% untouched. Because those accounts met the same criteria, any difference in performance can be put down to the programme.

Be patient with the revenue numbers. Jenna would give ABM a minimum of a year, partly because of the sales cycle and partly because of the change management it needs. Judging it at 30 days will not tell you anything.

Do this: agree three leading indicators and the control group in Step 2, set a baseline for each in week one, and report them monthly to the revenue leadership team.

Step 9: Your first 90 days, then the year ahead

Phase the rollout so each stage ends with a clear check before you add more. The sequence below is our suggested way to apply the steps above, not something Jenna prescribed, so adjust the timing to your sales cycle.

Expect the first quarter to be mostly about changing ways of working, which is why Jenna asks for at least a year before judging the result. Use the leading indicators to keep leadership confident in the meantime.

The Short Version

ABM is a way to put the right amount of time, effort and budget into the accounts with the most potential return, and a way to focus your reps' prospecting. It does not replace consistent calling, outbound and cold prospecting. Start small, build a sales process that gives reps consistent conversations with your market, write the rules down, and let those conversations teach you what to change.

If you want a second pair of eyes on how this fits your team and your stack, talk with Stakki about your stack: free advice, no pitch.