Skip to content

Company

We started with one uncomfortable question about every CRM we had used.

Why does a pipeline board look exactly as healthy on the day a deal dies as it did three weeks earlier? The stage had not changed. The buyer had stopped replying. Nothing in the software noticed, and nothing in the software cared.

How Macklore started

Every CRM we had worked in was excellent at recording process and blind to interest. It knew a deal was in negotiation. It had no opinion on whether anyone at the other company had opened a document in a fortnight. So reps ended quarters discovering that three deals they had counted on had quietly stopped moving weeks earlier, and nobody had been told.

The information was never missing. It was sitting in mailboxes, call logs and document analytics, spread across tools that did not talk to each other. A rep could reconstruct the truth about any single deal in about twenty minutes of searching. With thirty open deals, that is a working week nobody has.

So we built the thing that does the reconstruction continuously, on every deal, and then says something useful about it. Not a dashboard with another chart. A forecast on each card, a flag while the deal is still recoverable, and a draft that references the actual thread rather than a template.

The weather metaphor was not a styling decision. It came out of watching how reps talk about their own pipelines. Nobody says this opportunity has a declining engagement index. They say that one has gone cold. We built the interface around the language people already use.

Principles

Three commitments that decide what gets built and what gets refused.

They are short on purpose. A principle you cannot recite in a design review is not doing any work.

01

Score the engagement, not just the stage

Stage is an administrative fact about your process. Engagement is a fact about your buyer. Only one of them predicts whether a deal closes, and most CRMs only record the other.

How it shows up in the product

Every surface in Macklore shows both, side by side, and never blends them into a single comforting number. When they disagree, that disagreement is the most useful thing on the screen.

02

Suggest, never auto-send

The moment software starts talking to your customers unsupervised, you have traded a relationship for a metric. We are not willing to make that trade on your behalf.

How it shows up in the product

Macklore writes the draft and explains why. A rep reads it, changes what they want, and presses send. There is no setting that removes the person, and there will not be one.

03

Forecast from data, not gut feel

A commit list assembled from confidence is a wish list with a deadline. If a deal is a commit, there should be behaviour on the record that supports the claim.

How it shows up in the product

The engagement-weighted forecast exists so that a number can be challenged with evidence rather than with seniority. It sits next to your committed figure, not instead of it.

How we build

Four smaller opinions that shape the day to day.

These are the arguments that come up most often when a feature is being specified.

A CRM should earn its data entry

Every field a rep fills in by hand is a field the system failed to observe. We treat manual logging as a defect to design out, not a discipline to enforce.

An explanation beats a score

A number nobody can interrogate gets ignored in the first pipeline review where it is inconvenient. Every score in Macklore opens into the timestamped signals behind it.

Quiet software is better software

A product that raises thirty alerts a day has taught you to ignore it by Friday. Macklore stays quiet on the deals that are fine, which is what makes the flags credible.

Honest pipelines beat flattering ones

We price per seat rather than per deal so nobody is ever financially nudged towards deleting an uncomfortable record.

What we are working on next

Deeper account-level scoring for teams selling into several threads inside one organisation, and richer stage-dwell norms built from your own closed-won history rather than a generic benchmark.

What we will not build

An autonomous sender, a scoring model that rewards outbound volume, and any feature that makes a deal disappear from a pipeline without a person deciding it should.

Questions about the product, a rollout or a security review reach a person, not a queue.

Contact the team
Pipeline bulletin

Three principles, one engine, no autonomous sending.

If the pitch sounds right, the fastest test is your own pipeline on a Tuesday.

Import your open deals and read the first health pass. It takes an afternoon, and it tells you more about Macklore than any amount of reading about it.

  • Hexfield Interiors moved to at risk
  • Varun Steelworks flagged hot
  • 4 follow-ups queued for review