Connecting CRM, marketing automation and call tracking
How to connect your B2B tech website to CRM, marketing automation and call tracking properly so leads, attribution and routing all work end to end.
A B2B tech website that does not push clean data into the CRM, the marketing automation platform and the call tracking system is doing half a job. We see this constantly. Beautiful website, plenty of traffic, forms firing, but sales is working from a HubSpot list with broken UTM data, marketing cannot tell which content actually drives revenue and the SDR team is calling leads with no idea what page they came in on. Every part exists. Nothing is joined up.
Getting the integration layer right is unglamorous work, but it is one of the highest-impact technical projects in B2B marketing. Done well, it shortens response times, sharpens attribution and gives every team better data to act on. Here is how we approach it for the SaaS, MSP and IT service businesses we work with.
Start by mapping what you actually have
Before connecting anything, audit what you already run. The tech stack at most B2B tech firms has accumulated rather than been designed, and there is usually duplication.
We typically find some combination of:
- A CRM (HubSpot, Salesforce, Pipedrive, ActiveCampaign)
- A marketing automation platform (HubSpot, Pardot, Marketo, ActiveCampaign)
- A call tracking provider (CallRail, Mediahawk, ResponseTap)
- An analytics platform (GA4, sometimes Plausible or Fathom alongside)
- A CDP or warehouse if the business is more mature (Segment, Rudderstack, BigQuery)
- A schedule or meeting booking tool (Chili Piper, Calendly, HubSpot Meetings)
Map which tool is the source of truth for which object. Lead records: typically the CRM. Email engagement: typically the marketing automation platform. Anonymous web behaviour: typically GA4 or the CDP. Call data: the call tracking platform pushing into the CRM. Get this on paper before you wire anything up, because half the integration mistakes we see come from two tools both treating themselves as the source of truth for the same data.
Form integrations: do them properly the first time
The website-to-CRM connection runs through forms (and increasingly through chat). Get this right and everything downstream gets easier. Get it wrong and you spend the next two years deduplicating contacts.
A few patterns that work.
Use native form integrations where they exist
If you run HubSpot CRM and HubSpot Marketing Hub, use HubSpot forms. If you run Pardot with Salesforce, use Pardot forms. The native paths handle deduplication, lifecycle stage updates and field syncing in ways that custom integrations frequently break. Resist the urge to build a custom form just because the design team prefers a different look. Style the native form instead.
Capture UTM parameters and source on every form
Every form submission should include the original UTM parameters from the visit, the landing page URL, the referrer and the page the form was submitted on. Most forms only capture the page they live on, which means a paid ad that landed someone on a service page but converted them on the contact page records “/contact” as the source.
The fix is to use first-touch and last-touch hidden fields populated with cookies set on the first page visit. Most marketing automation platforms support this natively, but it has to be configured.
Validate emails and detect personal addresses
Add basic validation (rejecting invalid email format, optionally rejecting personal email domains for B2B forms) on the front-end and in the CRM. We use ZeroBounce, NeverBounce or the built-in validation in HubSpot. Personal email addresses are a flag, not a disqualification, but they should be visible to sales.
Push to enrichment
A work email and company name is often enough to enrich a record with industry, employee count, technology stack and revenue. We typically wire Clearbit, Apollo or HubSpot’s built-in enrichment to fire on form submission. This means sales gets a qualified record rather than a name and email.
Call tracking: the integration most B2B sites still skip
Call tracking is where MSPs, SAP consultancies and IT service businesses lose the most attribution data. The phone is a major conversion channel for these businesses. Without dynamic number insertion and call tracking, every phone lead is attributed to “direct” or whatever generic source they last touched.
A proper call tracking setup involves:
- A call tracking provider (we have used CallRail, Mediahawk and ResponseTap)
- Dynamic number insertion on the website that swaps the displayed number based on the visitor’s source
- Pool-based tracking for paid traffic so you can attribute calls to specific campaigns
- A connection back into the CRM so call records become lead records with source attribution
- A connection into Google Ads for offline conversion uploads, which feeds bidding algorithms
The setup is fiddly but only has to be done once. We routinely see businesses double their attributable lead volume after wiring this up properly, because all the calls that were happening anyway are suddenly visible. We talk about this in the broader paid context in conversion tracking for long sales cycles and attribution models for multi-touch tech.
Lead routing: do not make sales chase
A lead that takes 30 minutes to reach the right SDR has already cooled. We help clients build routing rules into the CRM so leads are assigned in seconds, not after a manual review.
Common routing dimensions:
- Geography (UK, EMEA, US, APAC)
- Company size (SMB, mid-market, enterprise)
- Product or service interest (which form they submitted, which page they were on)
- Account ownership (existing customer with an account manager, target account assigned to a named rep)
- Lead grade or score
For account-based routing, integrate with your account list or use HubSpot’s target accounts feature, Salesforce’s account hierarchy or 6sense / Demandbase if you run an ABM motion. We discuss the ABM angle in account-based ads on LinkedIn. For businesses with a customer login or partner area, the partner portal subdomain question often surfaces here too.
Closing the loop: revenue back to marketing
The integration most often missed is the one that pushes closed-won and closed-lost data back to marketing. Without it, marketing optimises for MQLs and pipeline, with no visibility on which channels actually drive revenue. With it, you can attribute closed revenue back to the original source, the original campaign and the original page.
For HubSpot CRM users this is built in. For Salesforce users, this means making sure the opportunity-to-original-source mapping is working and that closed-won deals trigger a sync back to the marketing platform. For sites with more complex stacks, a CDP or warehouse model is usually where this gets reconciled.
Once you have closed revenue mapped to source, GA4 conversion modelling and Google Ads offline conversions get dramatically better. The bidding algorithms learn from revenue, not just form fills, which usually delivers more efficient pipeline within a quarter.
Tag management and consent
Two technical pieces sit underneath all of the above.
Tag management: use a single tag manager (Google Tag Manager is the default) for all marketing tags, rather than hard-coding them in the site template. This gives marketing the autonomy to add and remove tags without engineering tickets, and it gives you a single place to enforce consent and load timing. For B2B teams running long sales cycles we increasingly recommend server-side tagging to harden tracking against browser restrictions.
Consent: GDPR and the UK equivalent require proper consent handling for cookies and tracking. Use a consent management platform (OneTrust, Cookiebot, Termly) that integrates with GTM’s consent mode, so non-consenting visitors do not fire marketing tags. This is also a WCAG and accessibility issue, as we noted in WCAG 2.2 accessibility.
Common mistakes we see
A few patterns to avoid.
- Building bespoke form integrations when a native one would do
- Capturing first-touch UTM but not last-touch (or vice versa)
- Leaving phone numbers as static text and losing all call attribution
- Letting sales bypass the CRM by using personal email and never logging
- Running two marketing automation platforms in parallel because nobody wants to migrate
- Treating attribution as solved by GA4 alone (it is not, particularly with iOS privacy changes)
Where to start if your stack is messy
Audit, simplify, then connect. We typically run a one-day stack workshop with the marketing and sales leaders, map the current state, identify the duplication and the gaps and produce a cleanup roadmap. Most B2B tech firms can consolidate their stack, plug the obvious gaps and improve attribution within a quarter.
This work also touches your website’s design and your SEO programme, because the form patterns, page structure and tracking implementation all sit on the same foundation. If you are scoping an integration project or replatforming and want to get it right this time, tell us about it. Our 30-minute calls usually flag a couple of obvious wins before any formal engagement starts.
Frequently asked questions
Should we use HubSpot's native forms or build custom ones styled to our brand?
How do we capture proper UTM and source attribution on every form?
Why does call tracking matter so much for MSPs and IT services firms?
More on Web Design
-
Web Design
B2B website migration without disaster: a replatform playbook
A practical playbook for B2B website migration covering planning, redirects, SEO preservation, content audit and the post-launch checks that protect rankings.
By Paul Clapp -
Web Design
The case for hand-coded websites in 2026
Why hand-coded websites built with modern frameworks like Astro outperform page builders for B2B tech, and where the trade-offs actually sit.
By Paul Clapp -
Web Design
Designing product pages that close enterprise deals
How to design B2B product pages that move enterprise buyers through evaluation, with the structure and trust signals that actually influence procurement.
By Paul Clapp