Shopping feeds skills: roles, tools and peak cover

What to take away

Decide whether to hire, train or contract before your next peak planning cycle, because shopping feeds work is splitting into data engineering, creative asset production and compliance checking.
Expect the strongest demand for people who can read a Merchant Centre diagnostics report and explain it to a merchandiser in plain language.
Budget for training as a line item, not a leftover.
Watch three triggers for a reviewa new sales channel, a change to image requirements, and any expansion into a regulated product category.
Use this guide to write role descriptions now, so recruitment does not stall in the weeks before peak.

Where the skills demand is moving

Three shifts are shaping the roles UK teams need: data work moving out of spreadsheets into pipelines, creative work absorbing the rules that govern product imagery, and compliance awareness spreading into everyday merchandising.

Data handling moves from spreadsheet to pipeline

Feed work used to sit with whoever was good at spreadsheets. That person still matters, but the volume of attributes and the frequency of updates now reward basic data skills: joins, deduplication, scheduled exports and error logging.

A useful test is whether someone can trace a rejected product back to the field that caused it. If they can, they can work a feed queue without escalating every item.

The tooling matters less than the habit, whether you run Channable, DataFeedWatch or scheduled exports stitched together in-house. Someone who logs a fix in the feed file, rather than keeping it in their head, stops a holiday turning into a rejection spike. Ask candidates how they hand work over.

Creative and compliance skills converge

The person who edits product images increasingly needs to know the rules those images must satisfy. Google's guidance on the image link attribute sets out size, format and content expectations that a designer can meet at source rather than after a rejection.

Compliance awareness is spreading into marketing roles too. ICO guidance on direct marketing covers the consent and contact rules around feed-driven campaigns, so any team that also runs email needs that knowledge in-house.

Roles and how to staff the work

Most UK shops do not need a large feed team. They need three capabilities covered: someone who owns the file, someone who owns the product data, and someone who can fix the joins when they break.

What a realistic team looks like

A mid-sized UK retailer with a few thousand SKUs might run one feed specialist, a part-time merchandiser and a contract data analyst during peak. The specialist owns the file, the merchandiser owns the product data, and the analyst fixes the joins.

Job titles and what to put in the spec

Shopping feed specialist:owns the feed file, runs scheduled exports, works the rejection queue and records fixes in the feed rather than in personal notes. Person spec: comfortable with spreadsheets or basic SQL, able to read Merchant Centre diagnostics, and able to explain a data fault to a buyer.
Product data merchandiser:owns titles, descriptions, images, categories and attributes. Person spec: product knowledge, attention to detail and familiarity with image link requirements.
Ecommerce data analyst (contract):builds joins, deduplicates records, sets up scheduled exports and error logging. Person spec: SQL or Python, experience with feed management tools such as Channable or DataFeedWatch, and clean handover discipline.

Mark each capability as hire, train or contract

There is no single market rate for feed skills, so build the plan from your own numbers: list each capability the team needs and mark it as hire, train or contract. The blank matrix below is a template you can copy into your planning document, then fill in your own hours and internal cost assumptions.

Twelve-month skills budget

Feed specialist£18,000
Training£1,200
Data analyst£4,000
Image support£2,500
Total£25,700

Compliance training is not optional

The FCA's finalised guidance on the fair treatment of vulnerable customers is written for regulated firms. Its themes of clear communication and avoiding harm apply to any marketing team that writes product copy. Feed descriptions are marketing copy. For non-regulated goods, the Advertising Standards Authority's CAP Code is the UK rulebook to train against, and Google's Merchant Center product data specification is the feed-level document for attributes, image links and data quality. In regulated categories such as supplements or financial products, the Food Standards Agency, MHRA or FCA rules apply before publication.

Triggers that should prompt a review

Reviews are usually forced by a change rather than a calendar date. Three triggers are worth watching closely.

Channel and rule changes

Adding a marketplace, or seeing a spike in disapprovals after a specification update, are both signals to check whether existing staff can absorb the work. If one person is the only route to a fix, that is a risk, not a skill.

The shopping feeds risk scenarios article sets out the failure modes that most often force an unplanned hire.

Category expansion and seasonal load

Moving into a regulated category, such as supplements or financial products, adds review steps that a generalist may not spot. Treat that as a training trigger before it becomes a compliance incident.

Seasonal load is easier to forecast. If your October queue is three times your July queue, plan the contract cover in advance rather than advertising in November. Book that cover in September and treat the handover notes as training material for permanent staff.

Common questions

These are the questions that come up most often when teams plan cover for feed skills.

Is a feed specialist a marketing role or a technical role?

It sits between the two. Strong candidates read product data and explain a technical fault to a buyer without jargon. Pure marketing backgrounds need data training; pure technical backgrounds need commercial context.

Should we train existing staff or recruit?

Train first where the gap is tool knowledge or attribute rules, because that is learnable in weeks, and Google's Skillshop hosts free training on Shopping campaigns. Recruit where the gap is data engineering, since it takes longer to build and is harder to borrow during peak.

How often should we revisit the forecast?

Twice a year suits most shops, plus an extra check when a channel changes its specification or you add a category. Tie the review to peak planning so the output is a decision.