A lead handoff two teams could finally agree on

When I joined Wowza, Marketing and Sales had been arguing about lead quality for long enough that neither team expected it to be settled. Sales said the leads Marketing sent were not worth working. Marketing said Sales was ignoring the good ones. Neither side could prove its case, because nobody had ever written down what a lead was. By the end, with the CMO and the CRO in the room to approve the decisions, both teams had committed to one set of definitions, every lead had a route, and the wait before Sales saw a lead had fallen by 75%, from roughly four days to one.

The company ran two go-to-market motions at once, a B2B sales funnel fed by a contact form and a product-led motion fed by trial downloads, both emptying into the same CRM. Rebuilding that CRM was already approved before I got involved. What had not been settled was the thing the rebuild depended on.

Key data points

• 75% less waiting. Leads reached Sales in about a day instead of roughly four, once scoring and routing replaced manual triage.
• 20% faster campaign delivery. Automating form setup and lead routing removed the step that had been slowing every campaign launch.

The Problem

Leads arrived from two places. The contact form produced Marketing Qualified Leads, or MQLs, someone who raised their hand through a marketing channel. The trial download form produced Product Qualified Leads, or PQLs, someone already using the software. Both landed in the same CRM and were treated as one kind of thing, although they behaved nothing alike: trial volume was enormous, mostly personal email addresses, with the same person often filling the form several times, while contact-form leads ranged from genuinely cold to ready to buy.

For example, someone would write in saying plainly that they wanted to buy, and nobody would answer for days. Not because anyone chose to ignore it, but because it sat in a queue alongside thousands of trial downloads that looked identical in the system.

From the Sales seat the funnel looked like noise, so the team worked its own leads and let Marketing's sit. From the Marketing seat there was no way to tell which leads had been worth anything, so nothing could be attributed to a channel or optimized toward. Both were partly right, and neither could prove it.

The Diagnostic

Everyone agreed the problem was lead quality. That was the symptom.

The actual problem was that nothing had ever been defined. Not what counted as a lead, not where Marketing's responsibility ended and Sales' began, not who owned which stage. The CRM was disorganized because the process was undefined, not the other way around, so rebuilding it alone was never going to fix anything.

A disagreement about quality cannot be resolved until both sides agree what quality means. Until then, the two teams are describing different objects with the same word, and every conversation restarts from zero.

Two of those objects had been collapsed into one. A PQL is evaluating the product already and mostly does not want a sales call yet, while an MQL signalled interest from outside and may be waiting for one. Treating them as a single queue meant the high-volume source drowned the high-intent source.

That produced the deeper accountability problem. Marketing was answerable for a figure it did not control, made mostly of leads it was never going to convert on its own, and Sales was answerable for working leads it had never agreed to accept. Neither team was behaving unreasonably. The process was.

So the work was not really a scoring model. The work was to manufacture an agreement, then build the mechanism that carried it.

The Approach

The CRM migration was approved a few weeks after I came into the project, with a HubSpot consultancy, Process Pro Consulting, brought in to design and implement it. Where I pushed was on sequence. I argued that the foundation was broken from the start, and that no funnel would work, however well it was built, until each stage had a definition, a handoff point, and an owner.

One constraint shaped everything. The SDR and BDR functions had been cut less than six months earlier; bringing them back was never on the table. There would be no human triage layer absorbing volume at the front of the funnel, so whatever we designed had to sort leads correctly on its own.

The first thing we built was not a workflow, it was an agreement. From there the funnel rebuild became the vehicle for the agreements, and every stage got argued to a definition somebody would commit to. The Director of Revenue Operations carried the view for Sales and held the final call, the campaign manager carried the business view for Marketing, and I carried the technical view, with the CMO and CRO approving decisions and the CEO signing off.

The definitions came next. MQLs and PQLs were separated, each with its own criteria and its own scoring, because they were being asked different questions. Then came the decision that mattered most for how the two teams behaved afterward: Marketing would be measured on MQLs, not on PQLs. PQLs were not abandoned, they had simply never been used well, so Marketing and Product took them on jointly with a strategy built for people already inside the product. What Marketing was accountable for became something Marketing could actually influence.

We added one path around the scoring on purpose. Some leads were not strong enough to hand over immediately but were worth testing, so they entered a nurture that read as a note from a salesperson asking whether they wanted to talk. A reply is a much better signal than a score, so any positive response skipped scoring entirely and went straight to Sales. It was worth giving up some consistency to catch intent a model would have missed.

The signature decision was the one that changed the working relationship. The definitions were fixed and non-negotiable, but the thresholds inside the scoring were deliberately left adjustable. If Sales felt the volume was too high or too low, nobody relitigated what a lead was. Marketing moved a threshold. A recurring argument between two teams became a setting one of them could turn.

The Built

With the definitions agreed, the build followed from them:

• Two lead definitions, with precedence. MQL and PQL defined separately, plus the rule for which one wins when a contact qualifies as both.
• The HubSpot Leads object and status model. Lead statuses introduced ahead of deal statuses in a CRM that had only used contacts and deals, giving Marketing accountable stages of its own and letting pipeline be traced to its source.
• Two scoring models. Separate logic per lead type, MQLs scored on behaviour as well as firmographics, with adjustable thresholds as the volume control.
• Routing automations and the priority-account exception. Workflows that routed each lead automatically, with a maintained list that pushed contacts from strategic accounts straight to Sales regardless of trial state.
• A nurture path with a bypass. A sales-voiced sequence where a positive reply skipped scoring, routed the lead to Sales, and booked the call automatically.
• Campaign templates and source stamping. Replicable form templates so Marketing could launch without a manual routing build, with source and form-fill history stamped onto the lead and carried to the deal.

Everything above shipped as a deliberate first version, and we knew it: the scoring was not sophisticated and leaned on heavy anchors, the priority-account list was maintained by hand, and the AI enrichment that gave Sales better context still depended on a person in Revenue Operations. Those were accepted starting points designed to evolve, not oversights. The funnel went live about a month before I left.

The Outcome

By the end, the two teams had stopped arguing about lead quality and started operating a funnel they had both signed. The change people felt day to day was not the scoring model, it was that nothing was ambiguous anymore.

• Leads to Sales 75% faster. From roughly four days of waiting, and in the worst cases no response, to same-day handling.
• Accountability each team could accept. Marketing owned lead statuses and MQLs, Sales owned deals, and neither was measured on something it could not influence.
• A threshold instead of a dispute. Volume complaints became an adjustment somebody made, rather than another debate about what a lead was.
• Campaign launches 20% faster. Marketing could ship by replicating a form template, without waiting on a routing build.
• A funnel built to be tuned. Definitions were fixed and the qualifying mechanisms left flexible, so the model could be adjusted as the team learned.

What had been a standing argument between Marketing and Sales became a set of definitions both teams agreed to, and a dial either of them could ask to turn.