HubSpot's Japan rebuild shows influencer ROI is a unit-of-measure problem

When HubSpot's marketing team reviewed its Japan automation programs, the automated emails were generating about 12 qualified leads per month against a goal of 160. The emails were going out and the workflows were running. Judged on activity, the program looked healthy. Judged on the unit the business cared about, it was delivering less than a tenth of its goal.

The obvious fix was a local hire. HubSpot says it could have hired a local marketing automationmarketing automationUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition → specialist in Japan, but instead rebuilt Japan's entire email automation program in two months, with zero additional headcount. HubSpot sells the CRMCRMCustomer Relationship Management: software and strategy to manage and analyse customer interactions throughout their lifecycle.View full definition → and automation software it used, so read the write-up as a vendor case study: self-reported, unaudited, and published partly to sell its product. The mechanics are still worth studying, because they match the problem CMOs now have with creator budgets.

My argument is this. Influencer ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition → past vanity metrics comes down to two decisions: what you count, and who owns the counting. Tooling comes after those. Most creator programs are where HubSpot Japan was: plenty of output, counted in units (views, likes, follower reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition →) that never add up to a business goal. The stakes are large. US creator advertising more than doubled from $13.9B in 2021 to $29.5B in 2024, and is expected to reach $44 billion in 2026, according to the IAB. A Cannes Lions 2026 session framed the gap bluntly: creators are in more than 60% of brand plans, yet fewer than 40% of marketers can prove the return.

How HubSpot rebuilt Japan's automation around qualified leads

HubSpot treated the problem as an organisational one. Wherever there's a primary market and a secondary one, the secondary market tends to lag, because the people with automation expertise are almost always scoped to the markets that generate the most volume. Japan had a target and no one with the skills to hit it.

The key decision was to split the job in two. HubSpot argues that you don't need one local specialist. You need an automation marketer who understands the technical side, and a local marketing or sales partner who owns the cultural validation. One person builds the machine and the other checks that it sounds right to a Japanese buyer. Neither is asked to do both.

The resourcing was small. By HubSpot's account, the work took about 12 days of automation marketer time in total (roughly 1.5 days a week over two months), 5 days of regional marketer time for QA, and no additional headcount. The output was 71 emails and 38 workflows built from scratch, organised into 2 core programs: one targeting prospects who hadn't engaged with sales yet, one targeting leads already showing interest.

The order of work mattered as much as the volume. HubSpot credits its decision to start with high-intent trigger-based emails as the reason it saw 9x growth in qualified leads in two months instead of incremental improvement over a year. Every message was set off by a behaviour that signalled buying intent. None went out just because it was on the calendar.

HubSpot's 9x qualified lead result, read with the caveats

The headline number is real but needs context. HubSpot reports that within the first reporting period, automation-sourced qualified leads jumped roughly 9x (from 12 to over 115 per month), and that it estimates that translates to about 44 incremental deals per month.

Four things a CFO would ask about:

• It covers one reporting period. We don't yet know whether the gain held over a full year.
• 115 is still short of 160. Coverage of the goal rose from 7.5% (12 ÷ 160) to about 72% (115 ÷ 160). That is a big improvement, and there is still a gap.
• The 44 deals are an estimate, not closed revenue. If they come from the roughly 103 extra leads (115 minus 12), the implied lead-to-deal rate is about 43%. That is high enough that you would want to see how it was modelled.
• The source is the vendor, writing about its own product in its own market.

None of this cancels the result. It shows the habit to copy: report the gap to the goal next to the growth multiple. The "9x" is the headline. The useful number is 160.

How do you measure influencer ROI beyond vanity metrics?

You choose one qualified outcome per creator program, set a target for it before the brief goes out, and judge every creator on cost per qualified outcome. Views and engagement stay in the report, but only as diagnostics. HubSpot's Japan case gives you the structure. The companies below show how far the industry has got in applying it.

L'Oréal, Unilever and The Independent show the range

L'Oréal has gone furthest on measurement. In partnership with WARC and Meta, it brought creators into the same full-funnelfunnelThe customer journey from awareness to purchase, typically Awareness, Interest, Consideration, Decision, Action, with prospects narrowing at each stage.View full definition → measurement framework as every other channel across 52 markets. The company has said the industry's reliance on easily accessible metrics such as likes, views and engagement rates is no longer sufficient. Circana, which sells marketing mix modelling and partners with L'Oréal, reports from the same work that branded and creator content together lift results by 35%. Two parties with a commercial interest stand behind that figure (Circana as the measurement supplier, Meta as the media seller), so cross-check it before you put it in a board deck.

Unilever made the biggest budget move. In March 2025 it announced a social-first strategy to concentrate 50% of marketing investment into social and influencers. Its 2026 first-half results report underlying sales growth of 4.8%, with 4.2% volume and 0.6% price, accelerating in Q2 to 5.8%. Those are company-wide figures, and they don't isolate what creators contributed. Trade observers say the same: as Unilever and L'Oréal increase their creator bets, the hard money trail to influencer-influenced revenue and profit contribution is far less evident. Unilever made the budget shift. The public evidence that creators drove the growth is still missing.

The Independent measures creator and social work in revenue. Its social-commerce revenue edged past its SEOSEOSearch Engine Optimization: the practice of improving your pages' natural (unpaid) rankings in search engine results pages to attract more organic traffic.View full definition →-based commerce revenue for the first time in September, CEO Christian Broughton told Digiday. Note that the company doesn't publicly break out the figures, so the milestone offers a directional signal rather than a measure of how far social commerce has pulled ahead. Even so, the publisher compares social and search in the same currency, pounds of commerce revenue. Most brands still don't do that with their creator spend.

New agencies are being set up around this gap. Adweek's framing of Cami Tellez's new firm Devotion was that social media has moved past the follower count, but influencer marketing has not. Tellez describes the scope as recruiting creators, gifting them, messaging them at scale, activating them, getting them to post, reconciling the content and paying them out, with AI agentsAI agentsAgentic AI refers to AI systems that pursue goals autonomously by planning, taking actions through tools, and adapting based on results, with minimal step-by-step human direction.View full definition → deployed across all of it. The step to watch is reconciliation. That is where a post gets matched to an outcome, and it is where most in-house programs fall apart.

The unit, trigger, two owners rule

Here is the decision rule, short enough for a whiteboard:

No creator spend without a qualified unit, a goal number, a high-intent trigger and two named owners.

• Unit. Name the one outcome that counts for this program: a sales-qualified lead, a first purchase, retail sell-through in a test region, a brand-lift score against a control group. Write the target next to it. If nobody can name the unit, the program is a media buy with no success test.
• Trigger. Start where intent already exists, as HubSpot did. That means creators whose audiences already visit your product pages, and amplification aimed at people who engaged with creator content. Broad-reach creator work comes later, once you know your conversion rates.
• Two owners. One person owns measurement: tracking links, codes, holdouts, and matching to yourCRM so creator-sourced contacts show up as records rather than impressionsimpressionsThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition →. A second person owns cultural fit: creator selection, tone, whether the content would embarrass the brand. Vanity metrics thrive when the creator manager does both jobs and grades their own work.
• Gap to goal. Report attainment against the target, never growth alone.

A worked example: why views rank creators wrong

Hypothetical, for arithmetic only. Two creators are each paid $30,000. Creator A delivers 1,000,000 views and 20 qualified leads. Creator B delivers 150,000 views and 60 qualified leads.

On a view basis, A costs $30 per thousand views and B costs $200, so A looks more than six times as efficient. On the qualified unit, A costs $1,500 per lead and B costs $500, so B is three times as efficient. The ranking flips completely. Then put those costs next towhat a qualified customer is worth over their lifetime, because $500 per lead is only cheap if enough of those leads convert at a margin that covers it. This is the HubSpot lesson at creator scale: change the unit and the budget moves.

Where a consumer brand's context differs from HubSpot's

HubSpot's case is B2B. Every prospect has a CRM record, and a qualified lead is a defined lifecycle stage. A shampoo brand sold through retailers has neither. For a brand like that, the qualified unit has to come from somewhere else: regional sell-through against a holdout, a promo-code redemption, or a modelled contribution like L'Oréal's.

Objectives differ too. The IAB found the top creator goals were building brand awarenessbrand awarenessThe degree to which your target audience recognises or recalls your brand, either prompted or unprompted. It measures how present your brand is in people's minds.View full definition → (43%), reaching new audiences (41%), enhancing brand reputation or trust (35%) and driving online sales or conversions (32%), while 40% of buyers rank overall ROI as their top KPIKPIKey Performance Indicator, a measurable value that shows how effectively you're achieving a specific objective, tracked over time against a target.View full definition → for creator campaigns. Awareness is a valid goal. It still needs a unit, such as lift against an unexposed control. Counting reach does not tell you whether the awareness spend worked.

The final difference is time. HubSpot's email triggers paid off within weeks. Brand-building creator work pays off more slowly, so set the measurement window when you set the goal, before results come in.

What to do on Monday with your influencer budget

• For each live creator program, write one line: the qualified unit, the goal number and the measurement window. Pause renewals for any program that can't fill in the line by Friday.
• Take last quarter's creator invoices, divide each fee by the qualified outcomes it produced, and rank creators by that cost. Put that ranking beside the ranking by views and take both to your next budget meeting.
• Name a measurement owner who doesn't manage creator relationships. Ops or analytics is fine, as long as it is not the person who picked the creators.
• Ask your agency or platform for order-level or code-level reconciliation of last quarter's posts. If they can only send screenshots of engagement, you have learned something about the contract.
• Choose one region or audience segment as a holdout for the next campaign, so you have a baseline to compare against.
• Move next month's test budget toward the creators whose content already sends people to product or pricing pages, which is the creator version of HubSpot's high-intent triggers.

HubSpot's most useful number is the goal of 160, more than the 9x. Write a goal number for a qualified unit into every creator brief this quarter, and views and likes go back to being diagnostics.