HubSpot Email Marketing common mistakes
HubSpot Email Marketing · Understand the traps and how to avoid them.
Treating a campaign's open rate as proof the email achieved the business goal, instead of defining a measurable goal (conversion, MQL, revenue) before choosing the send.
HubSpot reports open and click metrics prominently on every campaign, so candidates anchor on the most visible number and skip the goal-setting step that should precede it. Open rates are also inflated by privacy-related prefetching, making them a weak success signal.
What to do instead
For each scenario, write the goal first (e.g., 'generate 25 demo requests from a nurture segment') and the metric that proves it (form submissions, deals created). Then pick the send type, list, and CTA that serve that metric. If a question offers open rate as an answer to 'how should success be measured', reject it unless no downstream metric is available.
Designing a single 'blast' to the whole database for a scenario that describes different buyer stages or interests, rather than segmenting first.
Candidates default to the simplest send because segmentation feels like extra work, and they miss that HubSpot list criteria (properties, lifecycle stage, behavior) are the mechanism the exam is testing. Scenarios often bury the segmentation cue in a sentence about 'recently downloaded' or 'trial users'.
What to do instead
Before answering, scan the scenario for any differing attribute: lifecycle stage, product interest, engagement recency, geography. If two groups need different messaging, the answer involves an active or static list built on that property or behavioral filter, not a full-database send.
Confusing active lists with static lists and assuming an active list 'freezes' its membership at send time or that a static list updates automatically.
The names sound interchangeable in casual use. HubSpot active lists re-evaluate criteria continuously and membership changes as contacts qualify or disqualify; static lists only change when contacts are manually added or removed, or imported.
What to do instead
Memorize the rule: active = criteria-driven and self-updating; static = snapshot, manual or import-driven. Then check the question's wording - 'contacts who should stay on the list even after they convert' points to static; 'always includes anyone who meets the criteria' points to active.
Using only first-name personalization tokens and calling the email 'personalized', ignoring behavioral and lifecycle-based personalization.
Token insertion is the most visible form of personalization in the editor, so candidates stop there. Exam scenarios usually describe behavioral signals (pages viewed, past purchases, list membership) that should drive content blocks or send-time decisions.
What to do instead
Treat a first-name token as the baseline, not the answer. For personalization questions, look for the behavioral data available in the scenario and match it to a mechanism: smart content by list membership, dynamic product modules, or lifecycle-stage-specific copy. Choose the option that uses the specific signal described.
Sending a marketing email to a purchased or scraped list, or to contacts with no prior relationship, because the scenario frames it as 'growing the audience fast'.
Growth pressure in the scenario pulls candidates toward the fastest list-building option, and they underweight the consent requirement. HubSpot's terms and anti-spam law both require a legitimate basis for contact, typically opt-in.
What to do instead
Reject any answer that relies on purchased, rented, or scraped lists. The compliant path is opt-in collection (forms, landing pages, lead magnets) with clear notice of what the contact will receive. If an option promises volume without consent, it is wrong regardless of the metric it claims to improve.
Placing the unsubscribe link only in a footer image or omitting a physical mailing address, assuming the template handles compliance automatically.
Candidates see a polished template with a footer and assume legal elements are present. Images can be blocked, and a link inside an image is not a functioning unsubscribe mechanism. The physical address requirement is easy to overlook because it is not a clickable element.
What to do instead
Verify each email has a text-based unsubscribe link (and a preference center link where offered) plus the company's physical mailing address as plain text. When a question shows a template missing either, that is the defect to fix before sending.
Ignoring authentication setup (SPF, DKIM, DMARC) and blaming content alone when diagnosing deliverability problems such as landing in spam or being blocked.
Content and subject line are the visible parts of an email, so candidates jump to spammy words as the cause. Authentication failures are invisible in the editor but are a frequent root cause of filtering and rejection, and HubSpot requires connected sending domains to be authenticated.
What to do instead
Build a diagnostic order: authentication first (SPF/DKIM/DMARC records on the sending domain), then list quality (bounces, spam complaints, inactive contacts), then content. For any 'why are our emails going to spam' question, check whether an authentication answer is offered before choosing a content-based one.
Continuing to mail contacts who never open or click, instead of suppressing unengaged contacts to protect sender reputation.
Candidates equate a bigger send with a better result and worry that removing contacts shrinks the audience. In reality, repeatedly mailing non-engagers raises complaint and spam-trap rates, which damages deliverability for everyone on the list.
What to do instead
Use engagement recency as a list filter: build a segment of contacts with no opens or clicks over a defined window (e.g., 6–12 months) and run a re-engagement campaign, then suppress those who still do not respond. Choose the answer that reduces send volume to unengaged contacts rather than maximizing reach.
Reading a single campaign's metrics in isolation instead of comparing against a baseline, a control group, or an A/B test result before declaring a winner.
Dashboards display absolute numbers, and a 40% open rate sounds good without context. Candidates forget that seasonality, list composition, and send time all shift results, so only a controlled comparison supports a conclusion.
What to do instead
Ask what the comparison is: prior sends to the same segment, a holdout group, or an A/B variant with one variable changed. If a question asks whether a change 'worked', the correct answer references a controlled test or a baseline, not a raw metric. Reject conclusions drawn from a single send with no comparison.
Changing several elements at once (subject line, send time, and CTA) in an A/B test and then attributing the result to one of them.
Candidates want to improve results quickly and bundle changes to save time, not realizing that confounded variables make the result uninterpretable. They also test on samples too small to distinguish real differences from noise.
What to do instead
Isolate one variable per test, hold everything else constant, and ensure the sample is large enough to be meaningful before calling a winner. When a scenario describes multiple simultaneous changes, the correct critique is that the test cannot attribute the outcome - rerun with a single variable.