We use cookies to improve your experience on our website. By clicking "Accept all", you agree to the use of all cookies. Privacy Policy
Settings
BACK TO BLOG

The HubSpot Setup Mistakes We Fix Most Often

CRM & Customer Data
06 August 2026
·  Updated August 6, 2026
6min read
Nicolaas Kerkmeester
Nicolaas Kerkmeester
Director, Clear Click
A funnel of stacked navy bars with one outlined bar, representing the broken stage in a CRM pipeline
Table of contents
Heading
Heading

The five HubSpot setup mistakes we correct most often: lifecycle stages nobody agreed, dirty duplicate data, unowned automation, vanity pipelines and unused reporting.

Most HubSpot problems are not software problems. After years of inheriting portals in every state imaginable, the same five setup mistakes explain nearly every CRM that a sales team quietly stopped trusting. Here they are, with the fixes we apply.

1. Lifecycle stages nobody agreed on

Ask three people in the same business what a marketing qualified lead is and you will often get three answers, and a CRM configured around none of them. When stage definitions are fuzzy, every report built on them is fiction, and the sales and marketing argument about lead quality becomes unwinnable because both sides are right about different definitions. The fix is not technical: write the definitions down, get both teams to sign them, then make the portal enforce them. It is the first working session in every CRM setup we run.

2. Data nobody owns

Duplicate companies, contacts with four job titles, deals from 2023 still marked open. Dirty data is not cosmetic: automation acts on it, reporting aggregates it, and reps learn to work around the CRM rather than in it, which makes the data dirtier still. We assign every core object an owner, dedupe once, then build validation so the mess cannot quietly return.

3. Automation someone built and left

Most portals more than a year old contain workflows nobody can explain: sequences mailing a retired product line, lead rotation to someone who left, a webhook to a tool the company cancelled. Every automation should have a named owner and a written purpose, and anything that has neither gets switched off deliberately rather than left running out of fear. This discipline is most of what people mean by marketing automation done properly.

4. Pipelines that describe hope, not process

Eight deal stages with names like Engaged and Nurturing, where deals sit for months because no stage has an exit criterion. A pipeline should mirror the actual decisions a buyer makes, and each stage needs a definition of done. Fewer stages with honest criteria beat many stages that flatter the forecast. This is usually where sales process work starts.

5. Reporting built for the demo, not the Monday meeting

Portals often carry dashboards nobody opens because they answer no live question. We build reporting backwards from the decisions each meeting actually makes: what does the weekly pipeline review need to decide, and what is the shortest path to that number? Everything else is decoration.

The pattern underneath

Every one of these is a people-and-process agreement wearing a software costume. That is also why fixing them changes revenue rather than tidiness: a CRM the whole team trusts becomes the single view of the customer that marketing, sales and reporting all depend on. If your portal has drifted into folklore, ask us to take a look and we will tell you which of the five you have, and in what order to fix them.

Clear Strategy. Clear Growth. Clear Click.

Prefer to start with a conversation? Ask us for a second opinion on your current strategy. We will share where we would focus, and why.
Speak to Us