Compare order share, recipient share, and value share separately. Order share describes purchasing events; recipient share describes the distribution of recipient observations; value share describes spending concentration. None of these measures alone establishes a person’s preference. Label each denominator before using the data to choose a reward format.
A procurement team may ask, “Which gift card format do buyers prefer?” That question sounds straightforward, but the answer changes depending on what receives the weighting.
An order-weighted result treats every order as one observation. One order sent to a single recipient counts the same as one campaign sent to 500 recipients. This measure is useful when procurement is evaluating purchasing activity, administrative events, or the number of transactions a program creates.
A recipient-weighted result uses recipient observations as its unit. Confirm whether repeated recipients are deduplicated; the supplied aggregate does not establish a unique-person counting rule. A campaign sent to 500 people therefore contributes 500 recipient observations, while a one-person send contributes one. This measure is more useful when procurement wants to understand how many people actually receive each reward format.
The distinction matters because gift card programs often contain orders of very different sizes. A small branded-card order might serve a targeted sales team, while a large Choice Card campaign might cover a research panel or an employee population. Counting only orders could make the small targeted program appear more influential than it is across recipients. Counting only recipients could hide the fact that a particular format appears frequently in purchasing workflows.
Consider two illustrative formats:
An order-weighted view says Format A is more common, appearing in 80% of orders. A recipient-weighted view says Format B reaches 96.2% of recipients.
Neither result is wrong. They measure different units: purchasing events in the first case and people in the second.
Procurement should therefore label every chart clearly. “Share of orders” and “share of recipients” should never appear as interchangeable descriptions of preference.
Giftronaut internal statistics provide a useful example of why both measures belong in the same analysis.
Recipient-choice rewards account for 82% of Giftronaut recipients and 82% of total order value. These are two different denominators even though their rounded percentages match. They do not establish an average reward amount or a causal preference. This is a Giftronaut-specific internal statistic, not a market-wide preference claim.
Branded cards show a different pattern. They make up 31% of Giftronaut orders but only 18% of recipients. This pattern is consistent with smaller, targeted sends, as the source describes. It does not supply the underlying counts needed to report an average order size, and it does not establish why individual senders chose that format.
That comparison produces three distinct procurement insights:
The procurement conclusion should not be that one format is universally superior. Instead, the team should ask whether it is choosing a reward for broad recipient coverage, a targeted campaign, a known brand requirement, or a combination of these objectives.
A practical comparison should contain four layers: unit, audience, reward structure, and decision purpose.
Start by writing the denominator beside every metric:
The first two measures describe distribution. The third describes spend concentration. The fourth helps explain why order share and recipient share diverge.
Request the underlying counts before reporting average recipients per order. The supplied statistics establish a difference between shares, but do not provide a separate average order-size figure.
A recipient-weighted percentage is not automatically a survey of personal preference. It indicates how many recipients were covered by a reward type in the measured population.
Likewise, an order-weighted percentage is not automatically a buyer satisfaction measure. It indicates how orders were placed, which may reflect campaign rules, sender requirements, country availability, denomination limits, or a known brand requirement.
Procurement should use careful language:
That wording prevents operational behavior from being mistaken for stated recipient preference.
A blended global number can conceal important catalog differences. Giftronaut maintains separate local catalogs by country, and products issued for one country cannot generally be sent to recipients in another country. Giftronaut’s catalog evidence also distinguishes local gift cards from global prepaid options.
A procurement comparison should therefore segment at least by:
Do not pool programs with different purposes merely because they use the same provider. Document which groups are included, and explain why they belong in the comparison.
Use order-weighted data when the decision concerns:
Use recipient-weighted data when the decision concerns:
Use both when procurement is selecting a standard program for multiple use cases.
On a Giftronaut Choice Card, the recipient selects the brand at redemption. On a Branded Gift Card, the sender selects the brand at order time. Record that distinction before calling a selection a “recipient preference.” A sender-selected brand is not evidence that the recipient would have chosen the same brand from an unrestricted menu.
Identify who made the brand-selection decision, then evaluate whether the available data answers your procurement question.
A Choice Card also lets the recipient select their country at redemption. This can address a sender’s uncertainty about the country catalog, but the resulting order mix still reflects the program the sender chose to offer. Do not label every operational choice as an individual preference survey.
The following framework keeps the two weighting methods connected to business requirements.
| Procurement question | Primary measure | Why it matters | Follow-up check |
|---|---|---|---|
| How many recipients receive each format? | Recipient-weighted share | Shows audience coverage | Segment by country and program |
| Which format appears in more buying events? | Order-weighted share | Shows purchasing frequency | Review average recipients per order |
| Which format carries more spend? | Order-value share | Shows budget concentration | Separate reward value from fees |
| Do senders need a named brand? | Format and order-level review | Identifies fixed-brand requirements | Confirm recipient country |
| Do recipients span countries? | Recipient and country analysis | Tests catalog fit | Check local versus global options |
| Is the program large or targeted? | Recipients per order | Explains weighting gaps | Compare campaign sizes |
| Can recipients choose their reward? | Product-structure review | Clarifies experience | Document Choice Card or branded selection |
This matrix prevents one metric from carrying too much interpretive weight. For instance, a procurement team might observe that branded cards are common by order count and still choose Choice Cards for a large multinational campaign because the audience-weighted requirement is different.
Ask for the definitions before comparing percentages from different suppliers or reporting periods. Does a canceled order remain in the order denominator? Does a recipient mean a unique person, an email address, a reward, or a row in an order? Does value mean face value, funding deposited, or money including fees? An answer to one question should not silently answer the others.
Record unavailable definitions as unknown. If two providers use different units, present their results separately rather than ranking them in a single chart. A visually tidy comparison can still be invalid when the underlying populations differ.
The same person may receive rewards in several campaigns. Without an explicit deduplication rule, a recipient total should not be presented as the number of distinct employees reached. For your own analysis, preserve a stable internal identifier where your organization permits it and document whether repeated occurrences are counted. This is an analysis design recommendation, not a claim that a particular Giftronaut export includes every proposed field.
A durable procurement dashboard should preserve the original observation level. At minimum, retain order ID, recipient count, reward format, product, intended country, currency, order value, and campaign purpose where available.
The team should then review:
Avoid combining all programs into one headline number without showing the composition. A period dominated by one large campaign may produce a different mix from a typical month. Compare like periods and record major campaign changes before interpreting movement as a change in preference.
The dashboard should also distinguish observed behavior from stated preference. If the team wants to measure preference directly, it should collect a separate recipient question rather than treating redemption or order data as a survey response.
Before finalizing a comparison, procurement should ask suppliers and program owners:
These questions expose the most common comparison error: treating a convenient percentage as if it answered every procurement question.
Giftbit’s bulk gift card guide separates spreadsheet sends, connected-tool workflows, and API-based sends. It also distinguishes delivery and claim tracking. Those categories suggest useful questions about program frequency and observation units. They do not establish Giftronaut integrations or prove that order counts measure satisfaction.
The original comparison framework above combines those workflow questions with Giftronaut’s own aggregate mix statistics. The supplied figures do not state a measurement date or all underlying counting rules; request those details before using them as a matched benchmark.
Order-weighted and recipient-weighted gift card preferences should be treated as complementary measures. Order-weighted data explains how procurement activity is distributed, while recipient-weighted data explains how many people each reward format reaches.
Giftronaut’s aggregate mix illustrates why a purchasing-event measure can differ from an audience measure. Use that distinction to improve your comparison, while retaining the limits of the supplied data.
The strongest procurement process reports both measures, adds order-value share, segments by country and use case, and checks catalog constraints before drawing conclusions. That approach turns a potentially misleading “preferred gift card” question into a decision framework that reflects buyers, recipients, geography, and program purpose.
Order-weighted preference counts each order equally, regardless of recipient count.
It measures purchasing frequency rather than audience reach.
Recipient-weighted preference counts each recipient equally within the measured population.
It shows how many people receive each reward format.
Prioritize recipient weighting for audience coverage and order weighting for purchasing workload.
Use both when setting an organization-wide reward policy.
Orders can contain very different recipient counts. Frequent small sends may dominate orders without dominating recipient observations.
Record the definition as unknown until the data owner confirms it. Do not replace it with an assumption from another report.
That depends on the dataset’s deduplication rules. Confirm the definition before describing recipient observations as unique people.
Order counts and order values have different denominators. Measure value share separately when assessing spending concentration.