Specsavers plugged appointment data into media buying, here's how it worked

Delivering incremental eye tests is the core challenge for Specsavers, but how in a Modern Search world when advertisers all have access to the same technology, do you differentiate and deliver a competitive advantage? At MG OMD, we believed that the solution lay outside of these platforms and required a combination of business data, business knowledge and custom tools connected to buying platforms to enhance those capabilities. The primary objective for our Specsavers performance activity is simple - drive incremental eye test volumes, as efficiently as possible. Our partnership with Specsavers has allowed us to build a learning agenda rooted in business insight and prioritised using our proprietary media maturity model. Key learning questions like:

How can we use first-party appointment data to prioritise advertising investment to areas where availability exists, and pull back investment where stores are unable to fulfil demand?

Our measurement learnings tell us that our brand PPC term [Specsavers] delivers little to no incrementality but turning off our pure brand campaigns leaves us open to competitor threats. How can we protect our visibility against [Specsavers] but only when competitor threats are present?

Knowing product collections like Sunglasses and Anti-glare lenses are triggered by weather variables, can we ingest live weather signals to decide the activation of our Paid Search campaigns accordingly?

How can a business-like Specsavers drive competitive advantage online, when competitors all have access to the same tools and would be potentially running activity in a very similar fashion?

Without an unlimited budget, how do you drive incremental growth across a wide range of markets, which not only see consistent opportunity but also see large demand spikes, such as sunglasses during the March – July period?

STRATEGY

Working with our OMG central technology teams we workshopped potential custom solutions to answer our specific learning questions for Specsavers. It is thanks to the curiosity and tenacity of a truly collaborative partnership between agency and client that the fruition of these hypotheses has come to life via the suite of tools we coin SpecsTech. Our Live Appointment Feed (LAF) addresses our 1st learning question and is designed to meet one simple goal – promote a best-in-class service from Specsavers from the customer’s first interaction by helping them find the most suitable store appointment according to their availability and location. Our LAF allows us to meticulously manage demand across Specsavers' 900+ store footprint and was particularly pivotal in helping stores meet COVID restrictions during the pandemic period.

Undeniably valuable to the business yes, but particularly innovative in its ability to overcome Google’s platform limitations surrounding postcode level targeting and granularity of creative syncing opportunities to 1PD. Version 2 of our LAF has seen us address limitations of previous models through a low, medium and high availability bucketing architecture allowing us to plug gaps in coverage, tailor ad copy based on availability insight & leverage modern search technologies including auction time bidding & machine learning.

But perhaps the most impressive part of this SpecsTech solution is our ability to continuously iterate and improve it, allowing us to evolve with changes in the industry (modern search) but also with our client’s digital maturity. Our latest iteration has seen us ingest revenue data alongside availability data to enhance our bidding strategies to prioritise prospects in locations with high availability and who are also likely to convert higher revenue. As we work with Specsavers to unlock more data points, the opportunities to continuously evolve this piece of SpecsTech are endless.

SearchProtect is the SpecsTech solution created to address our 2nd learning question and was built following a fundamental learning from a Geo-Test that told us our PPC term [Specsavers] drove little incrementality to the business. Whilst this boasted a significant budget saving that we could invest into generic tactics that would deliver incremental volumes, it failed to address the other element of our Search Strategy – to protect our brand against competitors. Enter SearchProtect, a script solution that monitors the Search Engine Results Page (SERP) to identify when competitors are running across specific Brand terms – if identified, our technology automatically activates SearchProtect campaigns. This script mechanism is not new to the industry, but what makes it unique is the build flexibility that comes from an agency-first technology model – that is, a model that builds for client needs first vs. a technology-first model that prioritises accessibility and scale. It’s this flexibility that helps us overcome other product limitations such as a finite number of keyword inclusions, a rigid approach to script activation and a static rate card costing model.

LAF & BrandProtect are two pieces of significant SpecsTech, built for purpose and redefining our Paid Search strategy for Specsavers. But good things also come in small packages and our meteorological script, the solution to our 3rd learning question, is a testament to that. Specsavers’ business sales data tells us that we see a surge in Sunglasses sales not only in the spring and summer but generally when the weather is nice. The same data also surfaces a direct correlation between demand for anti-glare lenses when it gets dark outside. Our Meteorological script is the piece of SpecsTech that enables our weather-focused campaign activation. It does this using a script solution linked to an open weather API that dynamically pulls locations into a tactical campaign based on the decision logic set out:

For Sunglasses, the script collates cities experiencing nice weather into a campaign targeting sunglasses keywords and creative

For anti-glare, the script logic uses the same API to activate antiglare campaigns during dusk and de-active during dawn.