HubSpot’s retrospective identifies four advertising trends associated with 2016: experiential marketing, virtual reality (VR), live video, and influencer marketing. That is the accurate short answer. It is not an independently established ranking of the year’s biggest or most effective trends.
This is a historical guide, not current platform advice. It explains what the examples involved, separates reported claims from evidence confirmed by official or corporate sources, and gives editors a practical way to preserve metric definitions. The examples can show creative possibilities, but they do not by themselves prove sales, return on advertising spend, adoption, or comparative effectiveness.
The source under review is HubSpot’s retrospective on 2016 advertising trends. Its supplied metadata lists October 1, 2024 as the publication and update date, but that date was not visibly confirmed in the rendered article.
What were the four advertising trends in 2016?
The four categories named by HubSpot were:
- Experiential marketing: a branded in-person experience designed for participation.
- Virtual reality: an immersive viewing format.
- Live video: real-time video distribution through a platform.
- Influencer marketing: a creator or public-personality partnership used to reach an audience.
These are not four equivalent kinds of trend. Experiential and influencer marketing describe approaches to engaging or reaching people. VR and live video describe formats or distribution technologies. Classify a campaign by its principal mechanism before comparing its results.
| Category | What changes | 2016 example named by HubSpot | Metric caution |
|---|---|---|---|
| Experiential marketing | The in-person experience | Suicide Squad pop-up at SXSW | Attendance is not reach |
| Virtual reality | The viewing format | Yosemite VR film | A use case is not adoption |
| Live video | Real-time distribution | Taco Bell Friendsgiving | Viewers and views may differ |
| Influencer marketing | The creator partnership | Honest Tea’s #RefreshinglyHonest campaign | Participation is not business impact |
Attendance, views, viewers, impressions, reach, and consumer interest are different observations. They should not be combined into a single effectiveness score without defined units, periods, platforms, and aggregation rules.
Classify a campaign by what it changes: the experience, the format, the distribution, or the partnership. Compare results only after defining the metric and its period.
Experiential marketing: design an experience, then document the claim
HubSpot described Warner Bros.’ promotion for Suicide Squad at SXSW. GMR temporarily converted a downtown Austin tattoo shop into Harley Quinn’s Tattoo Parlor, using film-themed decoration, character lookalikes, and temporary and permanent tattoos. HubSpot reported more than 2,400 SXSW attendees and 35.4 million impressions on Twitter and Instagram.
Those figures remain reported claims rather than independently confirmed results in the available evidence. The supplied GMR campaign URL now redirects to a general work archive and does not display the activation or its metrics. The redirect is evidence that the original linked case study is unavailable at that address, not confirmation of the numbers.
Keep the measurements separate. Attendance is a count of people at an event. Impressions are platform-distribution observations, and the combined figure should not be relabeled as unique reach or added to attendance. An editorial record should preserve the source wording, platform, unit, reporting period, and evidence status for each observation.
Virtual reality: an immersive format, not proof of sustained adoption
HubSpot’s example was an 11-minute Yosemite film produced for the National Park Service centennial with National Geographic, Facebook’s Oculus team, and Felix & Paul Studios. It is a useful historical example of an immersive film project. It does not establish broad consumer adoption, sustained interest, or business results. The White House URL supplied for the video was not confirmed as a currently accessible record, so current availability on Oculus Rift or Facebook’s 360-degree platform should not be assumed.
HubSpot also cited more than $1.1 billion in VR and augmented reality investment “in 2016.” The linked TechCrunch report described more than $1.1 billion invested during the first two months of 2016, citing Digi-Capital. Preserve that time window. It is not a full-year total.
HubSpot cited consumer-interest percentages from a Nielsen Media Lab survey, but the original survey wording, sample, and methodology were not independently checked in this audit. Those percentages should therefore remain attributed historical claims rather than authoritative benchmarks.
A separate corporate record provides context: The New York Times Company’s 2016 annual report records the acquisition of Fake Love and HelloSociety for an aggregate $15.4 million. The annual report supports the acquisition claim, although the PDF was not directly accessible through the research tool. The acquisition does not prove broad VR adoption or the agency’s current operating status.
A historical investment figure is usable only with its original measurement window. Store “first two months of 2016” as a period field rather than allowing a summary sentence to turn it into a full-year benchmark.
Live video: separate platform milestones from campaign metrics
Facebook’s timeline needs a correction. Its official January 2016 announcement described expanding Live video to U.S. iPhone users, with Android availability still forthcoming at that time. Its April F8 announcement included the Live API for developers. April was an API milestone, not the initial consumer launch.
HubSpot reported that Taco Bell streamed a Thanksgiving-themed Friendsgiving feast and drew 150,000 viewers. The wording does not establish whether that means concurrent viewers, cumulative live viewers, or total views. Preserve “viewers” exactly unless the underlying source defines it.
HubSpot also reported separate Lowe’s figures for a Black Friday broadcast and views through promoted posts. Do not merge live audience, replay views, promoted views, reach, or impressions. Record each broadcast or distribution period as its own observation, including whether the distribution was organic or paid when the source provides that detail.
Influencer marketing: record the partnership and the evidence
Influencer marketing is a partnership approach in which a brand works with a creator or public personality to communicate with an audience. HubSpot described Honest Tea’s #RefreshinglyHonest campaign as involving seven Instagram-famous mothers sharing difficult parenting moments. Treat that participant count as reported rather than independently verified. The associated Instagram post is one example, not a complete campaign record or proof of the full roster.
Record participation, sponsorship evidence, audience response, and business outcomes as separate claims. A creator post does not by itself establish audience size, campaign scope, or commercial effect. HubSpot also discussed disclosure criticism involving Kardashian Instagram endorsements in 2016. That historical episode should not be used as a summary of current law or platform policy.
A practical evidence workflow for a historical trend brief
The following is a proposed editorial control, not a HubSpot feature, vendor integration, or documented automation template. It can be maintained in a spreadsheet or database. AI can suggest fields from supplied text, but a person decides whether a source supports the claim and whether its evidence status can be upgraded.
Keep the data at distinct grains:
- Campaign record: one row for the activation and its stable identity.
- Claim-source record: one row for one claim supported or repeated by one source and one retrieval.
- Metric observation: one row for one measurement, platform, distribution type, and period.
- Aggregate summary: a separate row only when the aggregation scope and metric semantics are explicitly defined.
A proposed claim-source identity can use hash(source_url + normalized_claim + retrieval_date). A proposed metric-observation identity should also include the campaign, metric, platform, distribution type, period start and end, and source claim ID. These are illustrative schema choices, not vendor-published fields.
{
"campaign_name": "Suicide Squad SXSW activation",
"claim_text": "HubSpot reported more than 2,400 attendees.",
"metric_name": "attendance",
"metric_value": "more than 2400",
"metric_unit": "people",
"platform": null,
"measurement_period": null,
"source_claim_id": "hash of source URL, normalized claim, and retrieval date",
"source_url": "https://blog.hubspot.com/marketing/biggest-advertising-trends-2016",
"evidence_status": "reported by secondary source",
"verification_note": "Original GMR case-study URL redirects and does not display the metric."
}Do not use a lookup-then-create check as the only duplicate safeguard when research jobs can run concurrently. Enforce the proposed identity with a database unique constraint or a transactional upsert. Retain separate records for later retrievals and sources instead of overwriting prior evidence. The same campaign can have multiple citations, broadcasts, platforms, and measurement periods.
| Trigger | AI job | Validation and destination | Fallback |
|---|---|---|---|
| A campaign claim is found | Suggest campaign, category, metric, and source fields | Human checks source support; save a claim-source record | Mark unavailable if the page redirects or is deleted |
| A platform date is asserted | Extract milestone wording and date | Compare with official Facebook announcements; save a timeline record | Label rollout scope uncertain |
| A live metric says “viewers” or “views” | Preserve wording and flag ambiguity | Store one metric observation with period and platform | Do not infer concurrency, reach, or unique people |
| An investment total is cited | Extract amount and time window | Check the source period before publication | Use qualified historical wording or omit the number |
Use AI for bounded extraction, not final adjudication. Deterministic rules are appropriate for URL checks, date formats, numeric units, and allowed evidence labels. Escalate inaccessible sources, undefined metrics, conflicting reports, and any proposed verified status to an editor.
- Does the accessible source support the exact claim?
- Are the metric wording, unit, platform, distribution type, and period recorded?
- Are redirected or unavailable sources labeled rather than treated as verification?
- Does the row grain prevent multiple citations or periods from being collapsed into one observation?
- Has an editor reviewed any claim labeled independently verified?
If you are mapping an editorial research process before automating it, see ConsultEvo’s services overview.
What the four trends do and do not tell us
Experiential marketing changes the audience’s in-person experience. VR changes the viewing format. Live video changes distribution. Influencer marketing builds on a creator partnership. HubSpot’s examples show historical creative possibilities and platform activity, not a comparative effectiveness ranking, proven sales impact, or current adoption.
Use the roundup as a lead for further research. A campaign should be called verified only when accessible evidence supports the specific claim, metric, platform, and period being published. Where the original record is missing or ambiguous, preserve the limitation instead of turning a reported example into a benchmark.