Lead follow-up: how to build a process that doesn't rely on someone remembering

A form gets filled in on a Tuesday afternoon. The notification lands in a shared inbox, three people see it, and each one assumes someone else has it. By Friday nobody has replied, and the prospect has booked a call with whoever answered first. That's what lead follow-up looks like when it relies on memory. Nobody did anything wrong, and nobody did anything at all.

This post is about the process and the automations behind it: who owns a lead, what happens and when, and how to know it's working. If you want the business case first, I've covered that in a separate ROI post.

Why does lead follow-up break down?

Lead follow-up breaks down because the CRM gets set up to store leads, not to act on them. Nobody decides who owns a lead, how fast they have to touch it, or what happens when they don't.

Then the business treats slow follow-up as a discipline problem and tells people to respond faster. That works for about a week. The automation is already in the tool. Nobody wired it to a person, a deadline and an escalation.

The research says most companies are living with this. When Conversica sent test inquiries to 100 mid-sized and large companies, one in four never responded at all, and about 35% gave up after one or two attempts. (Conversica sells follow-up software, so treat it as their research.) Harvard Business Review reached a similar verdict years earlier: most companies are not responding nearly fast enough.

It isn't laziness either. In Salesforce's State of Sales survey, reps said around 70% of their time goes on non-selling tasks. Asking those same people to also remember every callback makes the busiest part of the team the database.

When I open a CRM where "leads come in and nothing happens," the lead is usually there. Nobody owns it. It landed in a shared inbox or a queue instead of on a person. The second most common thing: the record exists, but the context didn't travel. The form asked what the prospect wanted, and that answer never reached the screen the rep looks at. So the rep either doesn't call, or calls and opens with "so, tell me what you're looking for."

The cost nobody counts

The most common mistake I see is treating slow follow-up as a people problem, then making marketing decisions on the damage. Leads go cold because nobody called for two days, so sales marks them unqualified. Marketing reads that data, decides the campaign brings in bad leads, and changes the targeting or kills it.

The leads were fine. The system let them die, and now the spend decisions rest on corrupted data.

What does a lead follow-up system need?

A lead follow-up system needs five parts, and each one runs on a trigger rather than a good intention. Miss one and the others carry the weight until they don't.

Context that travels with the lead

Every lead should arrive with the same core information: who they are, how they found you (source and campaign), and what they asked for. That last one has to be at the top of the record the rep actually opens. A fast call with the wrong information still kills the lead.

A named owner within seconds

A person, not a queue. The owner is a field on the record, so it can be reported on and it can trigger things. "The team" can't be accountable for a lead. A person can.

An escalation rule

This is the step that takes memory out of the system. If no contact is logged within your response window, the lead moves to the next available person and the manager hears about it. Holidays and busy days are covered automatically, because the lead reassigns itself.

A sequence with a defined end

"I'll chase next week" is a plan that lives in someone's head. A lead follow-up process writes the sequence down: which touch happens on which day, through which channel, and what each one is for. It also decides when to stop before it starts.

A daily exception list

Something has to notice the leads that slip through anyway. Not a dashboard someone has to remember to open, but a list sent every day to each owner and their manager: every lead with no logged activity past your response window, and who owns it.

How to follow up on a sales lead, step by step

If I were building a lead follow-up system from scratch for a small marketing and sales team, these would be the steps, in this order.

1. Measure your real response time. Pull 90 days of leads and compare when each record was created with when the first activity was logged. Look at the spread, not the average. An average of 45 minutes can hide a handful of leads that sat for eight hours, and that's where the money leaks.
2. Route every lead to a person in seconds. Every form submission creates a CRM record, assigns it to a named rep (round robin or by territory), and pings them. The ping carries the context: name, what they asked about, how they found you.
3. Add the escalation. If no contact is logged within 15 to 20 minutes, reassign the lead and alert the manager. Pick a window your team can actually hit, then hold it.
4. Run a sequence with an end. Alternate channels, give each touch a purpose, and decide in advance when the lead leaves the sequence.
5. Log every touch automatically. Calls, emails and replies land on the record without the rep typing them up. That's what makes the reporting possible.
6. Close the loop with a reason. Every lead ends as won, not proceeding, or no response and into nurture. Nothing sits in "in progress" forever.

What about the five-minute rule?

You'll see the famous benchmark everywhere: the old Lead Response Management study (2007, more than 15,000 leads) found the odds of reaching a lead fell 100 times when the first call came at 30 minutes instead of five.

I think it's overrated as a universal standard. It works well where the first business to call back usually wins the job, like moving companies or pet groomers. For plenty of businesses, hiring someone to sit on inbound all day doesn't make sense unless the math and the demand support it. Set a response window you can actually hit, and make sure the rep has the right context when they call.

A lead follow-up template you can copy

For a brand-new inbound lead, the exact number of touches matters less than the response window. Agree the window and the plan, then execute it inside that window every time.

Once a lead is further along, the sequence gets more structured. For a solar installer's quotes, I built a ten-day sequence with a touch every day, alternating call and email:

For deals that stall later in the pipeline, I use three touches over about two weeks: a personal check-in, a useful resource, then a close-the-loop note.

The same procedure had three rules the team doesn't break:

• Every open opportunity always has a next task scheduled.
• A verbal yes needs a deposit within 48 hours.
• No opportunity sits without a defined outcome: won, not proceeding, or no response and into nurture.

Which parts of lead follow-up should be automated?

Automate everything around the conversation, and keep the conversation itself human. That's the split that holds up in practice.

What to automate

An automated lead follow-up system handles the plumbing so the people can handle the talking:

• Creating the record from every form, with source and campaign filled in.
• Assigning the owner. Many CRMs can do this natively. HubSpot's workflows, for example, have a "Rotate record to owner" action for round robin or load-balanced assignment. Read the edge cases, though: HubSpot's own documentation notes that if no eligible user is available, no owner is assigned. That's exactly the gap the escalation and the exception list exist to catch.
• An instant confirmation to the prospect, so they know a person is coming.
• The alert to the rep, sent where the team actually looks. Email alerts get lost in the noise. Use Slack, Teams, text or WhatsApp.
• The next task, created with a due date at every step.
• The escalation when the window passes.
• Moving dead leads to nurture with a reason code, so the reporting stays honest.

This is the work behind connecting the systems: the form, the CRM and the sales team's tools passing the lead along without someone copying it by hand.

What to keep human

The first call, and the follow-up calls on a high-value sale. I'd rather the system nudge a rep to call than send another automated email. It feels more premium, and it's what people are buying from.

Where automation makes it worse

More automations aren't always better. At Ecoplex (more on them below), 28 automations built by different vendors were overwriting each other and sending bad data to sales. Every one of them was "working."

What about AI lead follow-up?

AI lead follow-up tools are getting a lot of attention. I don't use them for outreach. There's still a stigma around AI contacting people, and in B2B, where the whole game is trust built through relationships, that's fatal. Where I do use AI is underneath the team: an internal knowledge base built on Claude or ChatGPT, so the people doing the follow-up get accurate answers faster. We cover the wider question in our AI follow-up piece.

How many times should you follow up on a lead?

Enough to give the lead a real chance, and then stop on purpose. For most teams that means more attempts than they make today, inside a sequence with a defined end.

The data suggests most teams stop far too early. A Velocify study of almost 3.5 million leads found that 93% of converted leads were reached by the sixth call, yet 50% of leads were only ever called once. Outbound points the same way: RAIN Group's prospecting research puts the average at 8 touchpoints to land a first meeting.

What I push back on is the opposite failure: an open-ended "just checking in" every few days with no end. The lead never gets a status, so the pipeline fills with deals that died weeks ago.

I learned a lot of this before Maverick Marketer, working in-house at Mission Cloud. The SDR team was busy across email, phone and LinkedIn, but every rep ran their own approach. One prospect got a single email and then nothing for three weeks. Another got a cold call with no context.

We repositioned the SDRs as "Cloud Advisors" and designed one seven-touch sequence with the team: email, phone and LinkedIn, consistent spacing, a purpose for each touch, and no meeting ask on the first one. Every touch was logged, so we could see reply rates by step and channel. Outbound leads rose about 30% with the same team. (The seven came from an existing framework, not our own math.)

That was outbound, but four lessons carry straight over to inbound:

• Decide when to stop before you start. Without a rule, prospects either get dropped early or chased forever.
• Mix channels. Each one should reinforce the others, not repeat the same message three ways.
• Log every touch. Then a bad week can be traced to targeting, message or volume instead of guessed at.
• Write it down. One shared pattern makes coaching specific, and new hires start from the playbook.

Why do automated follow-ups stop firing?

Automated follow-ups usually stop firing because a field they depend on is blank, and the field that breaks most often is the owner. If the owner field is empty, nothing downstream fires: no alert, no task, no escalation, and the lead never shows up in anyone's view.

The close second is the source field. Underneath both is field drift. When every campaign adds its own form, the field names drift apart and the automations start writing to the wrong places.

The Ecoplex rebuild is the clearest example I have. Ecoplex, a solar installer in Ireland, had a sales team saying the wrong thing on roughly one in ten calls, referencing a service the prospect never asked about. Prospects went cold before the conversation started.

I talked to marketing and the reps first, then traced the path from ad click to the moment a rep opens the record. Over a few years, three or four agencies and several freelancers had each added a piece. There were 19 Webflow forms feeding 28 Zapier automations, some of them overwriting fields in Salesforce. Money was going into Facebook, Google and organic, and no source data reached the CRM at all.

The fix was a rebuild, not a patch: 3 forms, 3 automations, one field set agreed with sales, full UTM tracking, and a lead layout that puts the right context at the top. Bad-data calls went from roughly 10% to zero, and the rebuild took about three to four weeks.

The advice I give most often: before you add a channel or a sequence, check whether the field it depends on is actually populated.

How do you know your lead follow-up process is working?

You know it's working when a real test lead reaches the rep's screen and phone with everything right, and when your response-time numbers move. Being switched on isn't the same as working.

"The Zap ran" is not the test. The test is that the rep got the alert, the record shows the right service, the source is filled in, and the task is due. At Ecoplex we kept a dedicated test page for the Salesforce web-to-lead form and tested end to end across five lead types.

Run this check after any change to forms or workflows:

1. Submit a real test lead through every live form.
2. Check the record. Right service, source and campaign, at the top of the layout.
3. Check the owner. Assigned to a named person within seconds.
4. Check the alert. It reached the rep where they actually look.
5. Let it sit. Did the escalation fire when the window passed?
6. Check the task. Due, dated and owned.

The first step that fails is the one to fix. Fix the field underneath it before you touch the automation on top.

A few lead follow-up best practices keep it working between checks:

• Give every automation an owner and an alert for when it breaks.
• Send the daily exception list to owners and managers.
• Review response time weekly by rep, by source and by time of day. If response time and leads with no activity aren't moving, the system is on but not working.
• Use the rep report for conversations, not punishment. If someone keeps ignoring leads, the manager has a specific conversation instead of a general reminder. The prospect gets a call either way.

Where to start

Start with the measurement. Pull 90 days of leads, look at the spread between created and first touch, and find the leads that sat for hours. Then fix the step that let them sit, which is usually the owner or the field underneath it.

If you'd rather have someone build it with you, our Lead follow-up build covers routing, follow-up automation, and the field mapping underneath so every lead reaches a named owner fast. It's a One-time Project: fixed price once scoped, defined start and end, discovery built in. The price is quoted after discovery, because we don't quote before we've looked.

The easiest first step is a 30-minute call with me. Tell me where your leads go quiet, and you'll leave knowing whether it's a project or a partnership, and roughly what it would cost.

A follow-up process that needs someone to remember isn't a process. It's a hope with a CRM attached.