How the work developed

The on-the-job training plan set out digital-content production and e-commerce operations as its two intended main tasks. Neither became a main task during the placement, and the OJT record consequently marks both as unfulfilled. The placement I actually undertook was predominantly SEO-focused. Alongside the plan came a learning list covering SEO fundamentals and backlink work.

Reading about backlinks is a different exercise from opening Ahrefs Site Explorer (opens in a new tab) on a competitor and deciding which referring domains (show definition) deserve a closer look, so using the tools on live campaigns was where the learning material became usable.

Backlink opportunity research and forum or brand-mention research became the recurring responsibilities and accounted for most of my hours. Both continued in parallel with the project work described below.

The research had a consistent shape. Candidates were gathered, checked against the client's situation and recorded with the reason for keeping or dropping each one, so that the sheet carried its reasoning alongside the entry. Both areas have their own pages: backlink research and forum and brand-mention research. Answers arising from the forum work were posted on Quora.

I prepared title tags and meta descriptions (show definition) for PickleGO subpages. Those pages list products that change, so copy naming a specific model or promotion would become inaccurate. I wrote to the category and the buying decision, which keeps the description accurate as the range rotates.

To decide what each page should emphasise I looked at how people search for that type of equipment and at how competing retailers describe the same categories. The principles in the Google SEO Starter Guide (opens in a new tab) informed the drafts, particularly on descriptions that summarise a page honestly.

A hearing care provider was moving to a new multilingual site and a colleague had prepared the mapping between old and new URLs. My part was to check that work. I went through the mapped destinations, compared them with what was live and logged the entries that did not line up so they could be corrected before the migration.

This was the part of the placement where the work was verification of someone else's preparation. The full account is on the site review page.

PickleGO needed ActiveSG class information collated on a recurring basis, and class availability changes frequently. Doing it by hand took approximately twenty minutes each time, so I built the collation into a Google Sheets workflow driven by Apps Script (opens in a new tab) with browser-assisted extraction through Tampermonkey.

The completed workflow reduced the recurring update to approximately one minute. Server side access and a reader service were both tried and set aside during the build, and the final browser-assisted architecture is described on the automation page.

These timings are practical estimates and not formally measured figures.

The backlink research raised the same considerations repeatedly, so I consolidated them into a detailed working document setting out what to check on an opportunity and what ruled one out. That document was then used to have AI remove clearly unsuitable opportunities from a discovery list before I reviewed the remainder manually.

This was my own research workflow. The reasoning behind it is set out in developing repeatable research criteria.

The DTMS SEO material gave me the fundamentals for assessing backlinks. Repeated prospecting added commercial feasibility to that foundation. An opportunity could be relevant and still be impractical when its expected placement cost exceeded an SME budget.

The DTMS material also covered title tags, meta descriptions (show definition) and search intent. Those concepts became relevant when preparing metadata for PickleGO category pages whose product ranges changed, where the practical requirement was keeping the copy accurate as the range changed.

The Digital Commerce material connected to the PickleGO work, particularly the two livestreams. Finding common ground with the audience mattered when communicating product value, and real-time communication felt less comfortable to me than written research.

That livestream experience is also where the persuasion material from my communication module became concrete. Classroom teaching about relating a message to the audience is straightforward to agree with on paper. Having to communicate product value live, in an area I am still developing, showed me what it actually asks of the speaker.

Looking back at the ActiveSG work, the change management material describes something I did without naming it at the time. The workflow kept Google Sheets as the working environment, kept the refresh under manual control and kept the existing spreadsheet usability, while automating the repetitive collation. The later learning is that digitalisation has to fit the working process people already use.

The recurring difficulty in the backlink work was finding realistic opportunities for SME campaigns. Competitor backlink profiles produced useful starting points, and commercial constraints then removed many of the entries that had looked attractive. My response was to make the research more structured. The considerations that kept recurring were consolidated into the working document described above, which was then used with AI to remove clearly unsuitable opportunities before I reviewed the remainder by hand. That was my own way of working, and I cannot say what happened to it after my placement ended.

The ActiveSG collation was a different kind of problem. It took approximately twenty minutes each time and class availability changed frequently, so the effort returned week after week. My earlier thinking about automation had already made recurring service and API cost a practical concern, and this task was a situation where a workflow could produce a meaningful saving without one. The completed workflow reduced my update process to approximately one minute. The implementation detail sits on the automation page. What mattered here was that the saving repeats every time the sheet is refreshed.

The recurring research and the supporting project work required decisions about where deeper investigation was worthwhile. The backlink working document reduced repeated qualification work. The ActiveSG workflow addressed a separate recurring collection task.

Adapting to that mixture is also where most of my day-to-day workplace learning happened. The feedback I received on it is covered on the impact page.