Gseddy attributes more than €25,000 in revenue to email during its first month working with Tetsuo , a cycling-accessory brand whose case the agency published Aug. 27, 2026.
The case concerns recovery of an existing commercial relationship. The agency reports a database of 27,000 subscribers, a deliverability indicator moving from eight to 65 out of 100 and a 60% open rate. These figures describe different aspects of the channel. None establishes, by itself, how much additional profit the intervention generated for the retailer.
The store identifies ERIC Y JOEL, S.L., tax identifier B10623452, in La Eliana, Valencia, in its legal notice . Gseddy's original account places the work in a fourth quarter without specifying its year. This revision preserves the publication date and leaves the exact results period unknown rather than automatically assigning it to the final quarter of 2026.
A dormant commercial asset needs its own evaluation
The agency describes a Shopify store and a Klaviyo account with little previous activity. Its explanation combines contact cleaning, rebuilding automated journeys, content for a considered purchase and adjustment of the software plan. Several components therefore changed together. The available account cannot isolate how much of the result arose from each individual part of that intervention.
The initial 27,000-subscriber database does not establish that 27,000 people were reached in the first month. It may contain suppressed contacts, invalid addresses or people who received none of a particular campaign. Without eligible recipients, sends and deliveries, dividing revenue by the initial database would create an apparent performance measure that does not represent the result per actual recipient.
Recovery has a particular logic for a product requiring consideration. If someone remains interested but does not understand compatibility, installation or differences between options, a message can help resolve a question. If interest has ended, making the message longer does not restore the relationship. Selecting contacts is part of the commercial mechanism, alongside deciding what those contacts need to know.
The analysis must distinguish a buyer, a subscriber and a contact available in the platform. A purchase does not establish a preference for every future communication, and retaining an address does not prove it remains useful for sending. The size of a database can be both a cost and an opportunity, depending on its actual composition and the relationship behind its records.
What moving from eight to 65 means
The reported change is 57 points in the indicator. It does not mean the share of emails arriving in the primary inbox rose from 8% to 65%. A score combines signals under a tool's rules. Converting that score into an observed delivery probability would change the kind of measurement being described and would create a claim the case does not establish.
Klaviyo's deliverability-hub documentation says its score combines opens, clicks, bounces, unsubscribes and complaints across 30 days. The reviewed guide places scores from 50 to 74 in the “Fair” category. This is product context rather than a reconstruction of the historical configuration used by Tetsuo during the reported intervention.
The agency's description of a healthy account therefore expresses its assessment rather than an independent certification of inbox placement. The direction of the reported indicator is favourable, but assessing stability requires examining the components behind it. An improvement resulting from sending to fewer people differs from better response within the same observed population. Both can have value, but they need different explanations.
The recovery also creates a timing problem. A score calculated across a rolling window can include activity from before and after the intervention. Relating a single final score to every purchase in the month would require knowing the relevant dates and sends. The case does not publish a daily series that would allow the relationship to be examined in that way.
The team would need to observe bounces, unsubscribes and complaints separately, each with its own denominator. This proposed assessment could distinguish improved recipient selection from better message performance. It does not attribute undocumented instrumentation to Gseddy. It identifies the evidence that would help explain the reported recovery and the conditions under which the account's improvement could be considered persistent.
A 60% open rate needs context
A recorded open does not always identify human reading. Klaviyo explains in its deliverability guide that Apple's Mail Privacy Protection can prefetch a tracking pixel and inflate the measure. The case does not identify the filtering applied to its reported 60% rate or describe the email-client composition of the observed audience.
It also remains unclear whether that rate covers campaigns, automated flows or both. Messages triggered by a particular behaviour are observed at a different moment from general broadcasts. Comparing those groups without a common definition can turn a more interested recipient selection into an apparent improvement in content quality. The missing message-type breakdown prevents that explanation being tested against the public account.
For historical context, Klaviyo's sporting-goods figures for July-September 2023 show average campaign opens of 43.77% and clicks of 1.75%; flow figures are 56.73% and 6.28%. The guide includes Apple opens and account-selected industry categories. Its table does not identify a Spanish sample or a population directly comparable with Tetsuo's reported recipients.
These values help explain why campaigns and flows should be separated, but they cannot rank the brand against today's Spanish market. Their period is older, and comparable sample sizes and compositions are missing. Using them as a 2026 performance league table would suggest a degree of precision the available evidence does not offer. Historical context must remain clearly distinguished from current client results.
The useful question after an open is whether an interaction helped someone decide. A click to a technical explanation, a resolved query or a completed purchase describes a more specific step, although none proves causality alone. Content evaluation should observe those steps rather than optimise only the subject line to generate a recorded opening event that may not represent reading.
Spanish context also separates attention indicators from commercial outcomes. In the 2025 edition of Digital Response and Adigital’s study reported by Statista, 82.5% of surveyed professionals said they used conversions or sales to evaluate the channel, compared with 66.7% using opens. These percentages describe chosen metrics rather than achieved rates. For Tetsuo, the contrast prevents the reported 60% open rate from answering the money question: sales need their own definition, denominator and treatment of returns.
Opens are measured, but cannot stand in for sales
Metrics used to evaluate email programmes in Digital Response and Adigital’s Spanish study, 2025 edition.
Tap or hover for details. All values are also available in the data table.
Collapse notes
Survey of 114 professionals responsible for CRM, email or automation, according to the original report; CAWI according to Statista. Multiple answers. Report p. 17 matches the Statista chart: clicks 76.3%, rather than 76.5% in its summary. Metric use, not average performance or Tetsuo results.
Explore the data table
| Category | Respondents |
|---|---|
| Conversions and sales | 82.5% |
| Clicks | 76.3% |
| Opens | 66.7% |
| Customer lifetime value | 28.9% |
Attributing revenue does not establish additional revenue
The more than €25,000 is presented by the agency as email revenue. The account does not provide an order file, net revenue, costs or a comparison group. The figure therefore retains its stated scope: revenue attributed under an unspecified rule, rather than an independently estimated sum the business would have lost if the messages had never been sent.
Klaviyo's attribution explanation describes configurable windows and displays five days as default values for email opens and clicks. These explain how an available rule works. Tetsuo's settings are unknown, so this publication does not use the defaults to reconstruct the brand's result or infer which specific customer interactions received credit for the reported purchases.
Someone may discover a product through advertising, visit the store and later open an email before buying. An attribution rule can assign that purchase to an interaction without measuring which sales were additional. This example explains a measurement limit. It does not describe an observed journey of particular customers or deny that email can help someone make a considered purchasing decision.
The case compares the result with very little previous activity. That can make a neglected channel visible, but it does not create a reliable counterfactual. Reorganising communication and attribution rules can move credit from other channels. Assessing business growth would also require comparable total revenue and contribution, alongside the reporting label attached to each purchase in the marketing platform.
Returns introduce another limit. An initially attributed order can stop producing revenue or retain less margin after a change. Commercial evaluation needs a fixed point at which the outcome is treated as final, with the same rule applied to all groups. The case does not specify that treatment. Its reported amount is therefore not converted here into net profit.
Content should resolve a specific purchase question
A sequence for technical accessories can be organised around the questions needed to decide whether an option is suitable. More words do not improve the choice unless they explain a relevant difference. A proposed evaluation would identify the uncertainty each message addresses and the reasonable next step it offers, without inventing results or claiming an observed sequence beyond the agency's account.
Compatibility explanations need verified product information and a way to examine exceptions. Usage explanations can clarify when a feature matters and when it does not. Price communications need to show actual conditions. These serve different functions. A promotion should not substitute for the technical answers a buyer still needs before deciding whether the advertised accessory will work in the intended situation.
Interest signals also require caution. Viewing a product record does not establish immediate purchase intention, and responding to every visit with another message can become excessive. Before activating a journey, eligibility, signal validity and exit conditions should be defined. This is a design proposal; the source does not publish the detailed rules of the automated flows rebuilt by the agency.
A completed purchase should change the conversation. Continuing to send a message intended for someone who has not bought can contradict the customer's actual situation. Journey review needs to connect purchase and return events with communication rules. The relevant condition is the person's current relationship with the product, rather than the continued availability of an older signal in a marketing database.
Post-purchase content can help customers use a product and resolve uncertainty about delivery or setup. Its value needs appropriate observations, such as reasons for support contact or reported usage problems. Demanding the same immediate purchase rate from every message would misread its purpose. Nor should supposed contribution from service messages be added automatically to revenue attributed to promotional campaigns.
Sender reputation creates a commercial constraint
A recovered database can deteriorate if sending expands faster than recipient relationships. Sending capacity sits inside decisions about frequency, relevance and preferences. Opens from an initial campaign do not establish how long the same treatment can be repeated without increasing unsubscribes or complaints. Growth in communication volume needs to be examined alongside changes in recipient response and tolerance.
Gmail's sender guidelines require Postmaster Tools spam rates to remain below 0.3%. This is a provider's operational threshold, not a conversion rate or a measure of satisfied customers. The case does not publish this figure for Tetsuo, so the brand's position relative to that requirement cannot be confirmed from the available evidence.
The distinction matters because small volumes can produce unstable percentages. Evaluation should preserve the complaint count as well as the population used to calculate its rate. A marketing-dashboard rate need not exactly reproduce a mailbox provider's rate. Its definition and observation population must be identified before comparing those measures or treating one as evidence that the other requirement has been met.
When recovering contacts, reviewing preferences can be more useful than an immediate mass campaign. People should understand their relationship with the brand and the communications they receive. The proposed assessment checks that context and separates available contacts from eligible contacts, without assuming the entire historical database retains the same interest or the same scope of permission for marketing.
Evaluating contribution with a defensible design
A useful experiment could reserve a comparable group of eligible people without a particular promotional communication while maintaining necessary service for everyone. Assignment should precede observation of purchases and remain stable during the defined period. The comparison would include non-buyers, because excluding them would transform the result into a description of selected purchasers rather than the treatment's commercial effect.
The primary commercial outcome could be net contribution per assigned person. Measuring it would require net sales and costs attributable to the group under the same observation horizon. Clicks and queries would provide intermediate explanations. This is a proposed evaluation, rather than an experiment Gseddy or Tetsuo have announced in the documentation examined for this revision.
The horizon matters for a considered purchase. Too short a period can miss later decisions; an indefinite one can associate a message with distant purchases through coincidence. The choice should be made before reading outcomes and applied consistently across campaigns and flows. The design should also document how other promotional actions are prevented from contacting the reserved group and obscuring the comparison.
Seasonality deserves attention because the agency mentions a fourth quarter. Offers, availability and demand can change in that period, but the case's year is unknown. A cautious account keeps seasonality as a possible comparison factor. It does not attribute revenue to a particular Black Friday campaign or commercial calendar the source never identifies, even if such an explanation appears plausible.
Separating operational savings from commercial growth
Gseddy says it adjusted Klaviyo spending without publishing the amount saved. Reducing a bill has value even when sales do not grow. An economic account should keep that saving separate from additional commercial contribution and deduct the cost of the work performed. Combining several unspecified benefits into one return figure would make the project's actual profitability harder to evaluate.
Content production also has costs. Preparing technical explanations, reviewing segments and maintaining automation require continuing work. An account performing for one month does not show that costs remain unchanged as segments or the catalogue expand. Sustained commercial performance needs observation of relationship quality and of the resources required to maintain that quality beyond the initial period described in a case study.
Continuing work cannot be assumed from a software subscription either. In that study, 50% of professionals rated their organisation’s email mastery as intermediate and 39.5% rated it advanced. This is self-assessment without a capability audit, and it does not classify Tetsuo. Understanding Gseddy’s reported recovery requires asking which work remains established after the first month: maintaining eligible contacts, technical content and rules consistent with orders, alongside paying for the platform.
Reported email capability is not a channel audit
Two selected levels of email mastery rated by professionals about their organisations; Spain, 2025 edition.
Tap or hover for details. All values are also available in the data table.
Collapse notes
Digital Response and Adigital, 114 professionals; CAWI according to Statista. Organisational self-assessment, not examined knowledge or performance. Two levels shown rather than the full distribution: the elementary level differs between the original report’s chart and prose. Does not classify Gseddy or Tetsuo.
Explore the data table
| Category | Respondents |
|---|---|
| Intermediate mastery | 50.0% |
| Advanced mastery | 39.5% |
Coordination between store and platform introduces another condition: both systems should represent the same transaction. Order identifiers, status and net value should be reconcilable. Duplication or currency differences can appear as dashboard growth without existing in the commercial accounts. These problems have not been observed at Tetsuo; they are conditions that a reliable financial review would need to examine.
A promising recovery with measurement still outstanding
Public data describe an agency-reported recovery and a mechanism centred on audience, content and automation. They do not establish an independent inbox-placement rate, email's incremental contribution or net profit during the first month. Official product guides clarify how the measures should be read without supplying the missing client evidence or independently validating the agency's commercial claims.
This revision uses the original case, Tetsuo's own documents and official Klaviyo and Gmail documentation. It claims no nonexistent interviews or access to private accounts. Amounts and scores remain reported figures; the 2023 benchmark is identified as historical context. Exact intervention dates, the sent population and attribution rules remain unknown rather than being supplied through assumptions about the platform's defaults.
The next useful evidence would cover several periods of eligible and reached recipients, net orders and costs, alongside a comparison specified beforehand. It would show whether the first month represents a recovered relationship that continues generating value or an initial result requiring adjustment. For a technical purchase, the quality of customer guidance and the quality of its measurement need to develop together.
In this article
Marian Vasile
Marian Vasile is a MarTech and growth professional. His work spans product marketing, SEO, paid advertising and deep online tracking, including behavioral analytics and attribution.