Fourteen years of HubSpot work leave a mark on you. You stop being surprised by broken portals. You start recognizing the same problems almost immediately, the same warning signs, the same workarounds people build on top of a foundation that was never quite right.
This is not a list of theoretical mistakes pulled from a blog post. Each of the seven items below is something we have encountered at a real company, sometimes more than once in the same month. Some are obvious. A few are genuinely sneaky. All of them are fixable, but most become more expensive the longer they sit.
If you are using HubSpot and some of this sounds familiar, that is actually useful information. It means the problem is known, documented, and there is a clear path out.
Mistake #1: Lifecycle Stages That Nobody Actually Manages
This is the most common thing we find in a portal audit, and it never stops being a problem. A company sets up HubSpot with the default lifecycle stages, which are Lead, MQL, SQL, Opportunity, Customer, and Evangelist, and then uses exactly none of the logic behind them.
What happens in practice: contacts land in the portal and sit as Leads indefinitely. Nobody has defined what moves a contact to MQL. Nobody has agreed on what SQL means. Marketing is calling contacts qualified because they downloaded a whitepaper. Sales is ignoring MQLs because they know from experience that marketing’s definition of qualified and their own are two entirely different things.
The result is both a reporting problem and a relationship problem. Your funnel data is fiction. And your sales and marketing teams are working from different playbooks without either side realizing it.
We reviewed a portal for a SaaS company last year, where 84% of their contact database was still sitting in the default Lead stage. Contacts from 2019. The marketing team had no idea.
How to fix it
Before you touch any workflow, bring your sales and marketing leadership together and define each stage. Write down the specific criteria a contact must meet to move from one stage to the next. What action did they take? What property value changed? Who made the judgment call?
Once those definitions exist, build them into HubSpot as automated transitions for objective criteria and as manual checkpoints for human judgment. Then create a shared dashboard that shows monthly stage distribution so both teams can see the same data.
This single conversation, done properly, typically fixes more downstream problems than any technical change you could make.
Mistake #2: Workflow Spaghetti
There is a specific moment in many HubSpot portals when automation stops being a system and becomes a problem. It usually happens gradually.
Someone builds a workflow to handle a new campaign. Someone else adds one for a handoff process. A third party creates one to address a data issue. Nobody documents any of them. Nobody names them consistently.
Six months later, there are forty active workflows, and nobody is sure what half of them do.
We have walked into portals with over 150 active workflows, many of which overlapped, conflicted, or triggered contacts through sequences they were never supposed to be in.
In one case, a company unknowingly re-enrolled contacts in a sales sequence after they had already become customers because a workflow condition was checking the wrong property.
The trap with HubSpot automation is that it is genuinely easy to build. That is the platform’s strength. But easy to build does not mean easy to manage at scale, and most teams do not realize how much technical debt they are accumulating until something goes visibly wrong.
Automation debt is exactly like financial debt. You can carry it for a while without feeling it. Then something breaks at the worst possible moment, and the cleanup takes three times as long as the original build.
How to fix it
Start with a full workflow audit. Go through every active workflow and ask three questions: Is this still needed? Is it working as intended? Does it conflict with anything else? Archive anything that fails the first test. Fix anything that fails the second or third.
Then build a naming convention before you add anything new. A simple structure like Team, Function, and Trigger works well. Marketing, Lead Nurture, and Form Submit are readable, searchable, and tell the next person what they are looking at. Use folders. Create a workflow registry document in which each entry includes its purpose, owner, and last review date.
You do not need 150 workflows to run a sophisticated HubSpot setup. Most well-built portals we have seen run on 20 to 30 carefully maintained workflows, documented and actually understood by the people responsible for them.
Mistake #3: Lead Scoring That Nobody Trusts
Lead scoring sounds straightforward in theory. Assign points to actions and properties. Contacts with higher scores are more likely to buy. Surface those contacts to sales. In practice, most HubSpot lead scoring models fail within six months of being built, and the failure is almost always the same.
The model was built by a single person, usually a marketing manager, with little input from sales. It assigns points to things that appear to be buying signals but do not actually correlate with deal closings. Lots of points for visiting the pricing page. Points for opening three emails.
A high score for downloading a case study. None of those are bad signals, but none of them are worth much in isolation, and nobody validated them against actual closed deals before the model went live.
Sales looks at the MQL queue, sees a list of contacts who are clearly not ready to buy, ignores the queue, and never mentions it to marketing. Marketing continues sending contacts over. The disconnect widens.
In a portal we audited for a mid-market healthcare company, their top-scoring contacts had an average close rate of under 2%. The sales team had stopped checking the MQL queue altogether, and no one had told marketing.
How to fix it
Rebuild lead scoring as a conversation, not a configuration task. Sit down with three to five salespeople who close deals regularly and ask them to describe their most recent five contacts who became customers. What did those people have in common? What properties? What behavior?
Then look at your closed-won deals in HubSpot and find the behavioral patterns. Which pages did those contacts visit before converting? Which content did they engage with? How many touches happened before they became a real opportunity?
Build a dual scoring model that separates fit from intent. Fit score reflects the contact’s industry, company size, job title, and whether they match your ICP. The intent score reflects their activity: pages visited, emails opened, content downloaded, and webinars attended. A high fit and high intent combination is a meaningful signal. Either one alone often is not.
Review the model with sales every quarter. Adjust point values based on actual outcomes. A lead scoring model that improves over time is worth far more than a perfect model that nobody touches.
Mistake #4: Property Hoarding
HubSpot makes it very easy to create custom properties. Too easy, as it turns out. A new campaign needs a field to track something. A product team wants to capture a data point. An integration writes a property to store sync data.
Before long, a contact record has 80 custom properties, half of which are empty, a quarter of which nobody can explain, and most of which sales ignores entirely.
Property hoarding creates three real problems. First, it confuses the people who use the CRM daily. When reps open a contact record and see a wall of fields with no clear purpose, they fill in the ones they know and skip everything else. Data completeness drops.
Second, it makes reporting unreliable because inconsistently populated properties produce misleading analysis.
Third, it complicates automations because workflows built on properties no one maintains will misfire or stop working silently.
We recently audited a portal where 60% of the custom contact properties had not been updated for over 18 months. The team had been making decisions based on segment data that was essentially abandoned.
How to fix it
Run a property audit. Export all your custom properties and score each one on three dimensions: how often it is populated, when it was last updated, and which workflows or reports actively reference it. Properties that score low on all three are candidates for archiving.
Before archiving, check whether any workflow or list is filtering on that property. If something depends on it, fix that first. Then archive the property. HubSpot does not delete archived properties, so the data is not lost; it also cleans up the UI and reduces confusion.
Going forward, create a property request process. Anyone who wants a new custom property should be able to answer: what decision this data will inform, who will keep it updated, and what happens to contacts whose field is empty. That one question eliminates many unnecessary fields before they are created.
Mistake #5: No Handoff Between Marketing and Sales
A lead becomes an MQL. A workflow changes the lifecycle stage and creates a task for a sales rep. The sales rep opens the contact record, sees an email address, a job title, and a company name.
There is no context about why this person was qualified, what they engaged with, what problem they might have, or what the right conversation opener is. The rep cold-calls a warm lead and blows the opportunity.
This is one of the most common and most avoidable failures we see. Companies spend significant budget generating leads and qualifying them through marketing automation, then hand them off to sales with almost no context. The handoff is a lifecycle stage change and a task. Nothing more.
The cost is real. A poorly executed handoff on a warm lead is not just a lost deal. It is a lead who was already interested, already engaged, already spending time with your content, and then had a bad first experience with your sales team. They are significantly less likely to re-engage than a cold prospect who has not yet formed an impression.
The sales-to-marketing handoff is where more deal revenue quietly disappears than almost anywhere else in the funnel. The lead generation numbers look fine. The close rate doesn’t make sense.
How to fix it
Build context into the handoff. At minimum, every MQL passed to sales should include: what content or action triggered the qualification, a summary of the contact’s engagement history in HubSpot, the fit score and intent score if you are running lead scoring, and any notable company-level activity if you are using company scoring or intent data.
Automate as much of this as possible using internal notes, tasks with descriptions, or a HubSpot-native handoff summary populated by workflow actions. Some teams build a custom handoff property that summarizes the qualification reason in plain language, which the workflow populates automatically when the contact reaches MQL status.
Then create an SLA. Marketing commits to passing contacts that meet the agreed-upon MQL definition. Sales commits to following up within a defined window, typically 24 to 48 hours for inbound leads. Both teams track their side of the SLA on a shared dashboard. The accountability changes the dynamic.
Mistake #6: Data That Never Gets Cleaned
Every HubSpot portal accumulates data problems over time. Duplicate contacts are created when the same person submits a form with a slightly different email address. Contacts imported from a legacy CRM have inconsistent formatting.
Properties populated by an integration that stopped syncing six months ago but left bad values behind. Company records with no associated contacts. Contacts with no associated company.
Dirty data is a slow problem. It does not usually break anything dramatically. It just gradually makes everything less reliable. Reports show slightly wrong numbers. Segmentation produces slightly wrong lists. Workflows fire on slightly wrong contacts.
The accumulation of small errors reaches a point where leadership stops trusting the data entirely, and that is when the real damage occurs.
We worked with a financial services firm last year that had been running HubSpot for three years without any data governance.
When we audited the portal, nearly 30% of their contact database was duplicates. Their monthly pipeline report had overstated the number of active opportunities for over a year because duplicate contacts were counted as separate leads.
Bad data does not just hurt your reports. It hurts your team’s confidence in the system. Once people stop trusting the CRM, they stop using it properly, and the data gets worse faster.
How to fix it
Clean up is a project. Maintenance is a process. You need both.
For the cleanup, run HubSpot’s native deduplication tool on both contacts and companies. Then use a more granular approach to catch near-duplicates that the native tool misses.
Go through company records without contacts and either associate them or archive them. Audit your integration sync logs to find properties being overwritten with bad values.
For maintenance, you should establish a data governance cadence. Monthly is usually right for most teams. Assign someone ownership of data hygiene as a defined part of their role.
Create a set of HubSpot reports that automatically flag data quality issues: contacts without email addresses, contacts without lifecycle stages, company records with no contacts, and properties that are unexpectedly empty on contacts that should have them populated.
Treat data hygiene as infrastructure, not housekeeping. The reliability of everything else in your HubSpot setup depends on it.
Mistake #7: Building for Features, Not for How the Business Works
This is the most systemic mistake on the list, and in some ways the most consequential, because it is the foundation from which the other six mistakes often grow.
HubSpot is a very capable platform. It can do a lot. And when a team gains access to it for the first time, or when a new admin comes in and starts exploring what is possible, there is a natural pull to build things because the platform can support them. New pipelines. New automation. New dashboards. New reporting structures.
The problem is when those things are built around what HubSpot can do rather than how the business actually sells, markets, and serves customers. We see this most often in three areas.
Pipelines with too many stages. The default HubSpot pipeline has deal stages that are generally applicable but may not reflect the specific steps in a company’s sales process.
Teams add stages to mirror every internal conversation, and deals progress through 10 or 12 stages with no clear exit criteria for any of them. Forecasting becomes impossible.
Dashboards that look impressive but do not inform decisions. Reports are built because HubSpot can produce them, not because anyone asked for them or will act on them.
Leadership gets a weekly email with 15 metrics, reads none of them carefully, and asks for different data at every quarterly review.
Automation that mirrors the org chart instead of the customer journey. Workflows are structured around internal team handoffs rather than around what the customer is actually experiencing.
The result is a series of automated touchpoints that feel disconnected from the customer’s perspective, because they were designed from the inside out.
The most expensive HubSpot setups we have ever fixed were built by people who knew the platform very well. Technical HubSpot skill and strategic business sense are not the same, and a portal full of sophisticated mistakes is harder to fix than a simple setup with a few gaps.
How to fix it
The fix here starts before you open HubSpot. Spend time mapping your actual customer journey: how do prospects find you, what happens during evaluation, where does the sales conversation begin, and what does the post-sale relationship look like? Then build HubSpot to reflect that map.
For pipelines, you should keep stages to the minimum number that reflect genuine decision points in the customer’s process, not internal milestones.
A well-structured pipeline for most B2B companies has five to seven stages. If you need more, that usually indicates some of your stages are internal process steps rather than customer journey stages.
For dashboards: start from questions, not metrics. What decisions does leadership need to make monthly? What does a sales rep need to see every morning to know where to focus?
Build only the reports that answer those specific questions. You can always add more later when someone asks a question you cannot currently answer.
For automation: walk through every workflow from the perspective of the contact it affects. Is the timing right? Is the message relevant to where they actually are in their journey? Would you want to receive this if you were them? If the answer to any of those is no, the workflow needs to be rethought.
How The Automation Strategy Group Can Help
The Automation Strategy Group has been working inside HubSpot portals since 2011. We are a certified HubSpot Solutions Partner with additional certifications across ActiveCampaign and Eloqua.
Our clients are B2B SaaS companies, healthcare organizations, financial services firms, and education businesses that want their HubSpot setup to reflect how their business runs.
The seven mistakes above are not abstractions for us. They are issues we have repeatedly fixed for companies at every stage of growth.
Some came to us with a brand-new HubSpot instance that needed a solid foundation. Others came to us after years of accumulation, ready to clean things up and start building the right way.
Book a free strategy session to talk through what you are working with or run a Blueprint 360 Audit, our structured HubSpot health check that identifies exactly where your portal has gaps and what to do about them.
Frequently Asked Questions
How do I know if my HubSpot portal has any of these problems?
The most reliable indicator is whether your sales and marketing teams agree on the pipeline data. If they argue about numbers at the monthly review, something structural is usually wrong. Other signs include reports that require manual adjustments before anyone can present them, sales reps who maintain their own spreadsheets alongside HubSpot, and workflows that no one is confident editing. A formal portal audit, whether conducted internally or by a consultant, will quickly surface the specific issues.
How long does it take to fix a portal with multiple of these problems?
It depends significantly on the size of the database, the number of active workflows, and the number of integrations in use. A focused cleanup of lifecycle stage logic and workflow documentation typically takes two to four weeks of dedicated work. A full portal rebuild, covering data structure, automation architecture, reporting, and team alignment, usually runs six to twelve weeks. The longer these problems have been accumulating, the longer the cleanup takes, which is the strongest argument for addressing them early.
Will cleaning up workflows and properties cause anything to break?
The right approach is to audit before you archive anything, checking which lists, workflows, and reports reference each property or workflow before removing it. HubSpot’s archive function is also reversible for most object types, which provides a safety net. That said, cleanup work on a live portal requires careful sequencing. If you are not confident in the process, working with a consultant who has done it before significantly reduces the risk of unintended consequences.
My sales team does not use HubSpot consistently. Is the CRM the problem?
Low adoption is almost always a symptom of a setup problem rather than a resistance to technology. If the CRM does not reflect how reps actually sell, if the fields they are asked to fill in feel irrelevant, or if the data they find when they log in is not useful to them, adoption will be low regardless of how much training you provide. The fix is usually to simplify the rep experience: fewer required fields, deal stages that match the actual sales conversation, and views that surface the information reps actually need to do their jobs.
Can we fix these problems ourselves, or do we need outside help?
Some of them are absolutely fixable internally, particularly property cleanup, workflow documentation, and lifecycle stage alignment. Those are more about discipline and process than technical expertise. The harder ones, specifically rebuilding lead scoring, restructuring data architecture, and redesigning reporting around real business questions, benefit significantly from an outside perspective. A consultant who has seen how these things work across many portals brings a reference point that internal teams typically lack.
Is it too late to fix the foundation if we are already live with HubSpot and have a full database?
It is never too late, though earlier is always better. We have rebuilt the data architecture and automation logic for portals with hundreds of thousands of contacts without losing data or disrupting active campaigns. The process requires careful planning and a phased approach, but it is very much doable. The companies that wait the longest tend to have the most painful cleanups, but they also tend to see the biggest improvement in pipeline reliability and team confidence once the work is done.
What is the first thing we should do if we suspect our HubSpot setup has problems?
Start with lifecycle stages. Check what percentage of your contact database is sitting in each stage, and then ask whether those numbers reflect reality. If most contacts are in Lead with no clear logic for how they got there or where they should go, that is your starting point. It is the structural issue that most commonly cascades into the other problems on this list: bad scoring, broken automation, unreliable reporting, and a sales team that has quietly stopped trusting the data.