Getting started with HubSpot CRM is less about buttons and more about order. The teams who succeed don't rush to import contacts — they set up their data model, pipeline, and users first, so the CRM reflects how they actually sell before a single record goes in. Skip that, and you spend the next year cleaning up duplicates, fixing deal stages, and re-teaching the team. Here's the practitioner's read on the right way to begin.
What should you set up first in HubSpot?
Set up your foundations before your data: account settings, your team, and a clear picture of your objects and properties. The temptation is to import contacts on day one because it feels like progress. It isn't — it's the step that's hardest to undo. Start instead by confirming your account basics (company info, branding, default currency, time zone), adding your users with scoped permissions, and deciding which objects you'll use: contacts, companies, deals, and tickets, connected by associations. HubSpot organizes everything as records linked together, so knowing how a contact relates to a company and a deal before you import is what keeps your data connected rather than orphaned later.
How do you set up your deal pipeline correctly?
Build your pipeline stages to match how your team really sells — and keep them few, clear, and based on buyer actions, not internal feelings. Your pipeline is the spine of your sales reporting, so it's worth getting right early. Map the real steps a deal moves through, and define each stage by something observable: "demo booked," "proposal sent," "verbal yes" — not vague states like "interested" that two reps will read differently. Fewer, sharper stages beat a long ladder nobody updates consistently. Worked example: a team replaced an eight-stage pipeline full of overlapping labels with five action-based stages; within a month their forecast got more accurate, because every rep now moved deals on the same triggers instead of guessing.
When should you import your contacts?
Only after your data model and properties exist — and only with a clean file matched on a unique identifier. The import is where early shortcuts turn into permanent problems. Before you import, make sure the properties you're bringing in already exist in HubSpot with the right field types, clean your source file (consistent formats, one property per column, deduplicated), and match on a unique key — email for contacts, domain for companies — so HubSpot updates existing records instead of creating duplicates. Import contacts and companies together with a shared company-name or domain column so they associate automatically. Then validate a sample afterward: did associations form, did properties land in the right fields, are owners correct? A clean first import is the cheapest insurance you'll ever buy on a CRM.
What habits keep a new CRM clean?
Decide your data-entry standards and required fields up front, because a CRM decays the moment people start entering data their own way. A new portal is at its cleanest the day you launch it; from there it only stays clean if you make clean the path of least resistance. Standardize key properties as dropdowns instead of free text, mark a small set of required fields so critical data is never blank, and agree as a team on what "good" looks like for a contact, a company, and a deal. Pair that with light automation — a workflow to set lifecycle stage, rotate new leads, or flag records missing key data — so hygiene runs in the background. The goal isn't a perfect portal on day one; it's a portal that stays trustworthy as it grows.
The IV-Lead take
The biggest mistake we see with new HubSpot accounts isn't a wrong setting — it's the order. Teams import first and design later, then spend a year paying interest on that decision in duplicates, broken reports, and a pipeline nobody trusts. Flip it: model the data, build the pipeline around how you actually sell, add users with real permissions, then import clean and validate. It feels slower for a week and saves you a year. A CRM is a system, and systems reward the people who set the foundation before they pour in the data.
Starting fresh on HubSpot, or inherited a messy one? Book a 30-minute portal audit — we'll tell you straight what to set up first and what to fix before it costs you. For the bigger picture, see how we approach HubSpot implementation and optimization.
Frequently asked questions
What's the very first thing to do in a new HubSpot portal?
Confirm account basics (company info, currency, time zone), add your users with scoped permissions, and decide which objects and properties you'll use — before importing any data. Setting the foundation first prevents the rework most teams face later.
How many deal stages should my pipeline have?
Few enough that every rep updates them consistently — often five to seven. Define each stage by an observable action, like "proposal sent," rather than a vague feeling like "interested," so your forecast stays accurate.
Do I have to import all my data to start using HubSpot?
No, and you shouldn't rush it. Get your data model and properties in place first, then import clean on a unique identifier. A small, well-structured start beats a large, messy one.
How do I keep the CRM clean after launch?
Standardize key properties as dropdowns, require a few critical fields, agree on data-entry standards as a team, and use light automation to set stages and flag missing data. Make the clean path the easy path.
