I often get asked about marketing strategies incorporating AI and data.
Here are four ways I am leveraging AI and data to boost performance for our clients, which you should consider integrating into your planning, whether from an execution or strategic standpoint:
1. Using SEMRush’s domain overview tool and ChatGPT to perform keyword research for search campaigns.
Before I create any search campaign, I locate the domains of competing brands. Once I compile the domains, I’ll input them one by one into SEMRush’s domain overview tool. Once input, I will click on “paid search traffic.” Once redirected, I will export a CSV of all paid keywords. Once I retrieve the keywords, I’ll engineer a prompt like, “You are an experienced Google Ads manager at an agency. Please organize these keywords into ad groups.” From here, I identify the keywords my competition is bidding on, ensuring I cover the basics and supplement with any necessary long-tail or KSP-specific keywords. (Want to improve your ad rank? The same tip works for determining keyword density in advertisement copy. Just use SEMRush’s ads copies tracker and engineer an ask that measures for keyword density. Of course, you could also use a COUNTIF function in Excel, but let’s work smarter, not harder.)
2. Using ChatGPT to build schema.
Want to develop model schema? Take a location detail page of someone ranking well. Import that URL into Schema validator. Identify the schema being used and prompt an ask to answer commonly-held information. (You’ll need to fine-tune your prompts and ensure clarity, but if you’re not proficient in JavaScript, ChatGPT generates it for you. This saves you time and effort. Once you have a working template, duplicating it becomes much smoother.)
3. Enhance omnichannel measurement by connecting a first-party identifier from your TV provider’s live exposure files.
Our TV partner sends live exposure files to a DMP. If you can generate a communicable first-party ID, you can join the data to identify overlap. The data can provide valuable measurement insights. Independently or separately, these datasets can be imported into a DSP in a privacy-compliant manner. In one particular use case, we observed a 17% increase in conversion lift post-advertisement within a 90-day conversion window. (Almost a lawyer, advisable to independently validate all agreements with your company’s legal counsel.)
4. Use non-endemic 1P data from RMNs to enhance your understanding of customer behavior.
You can activate data from a RMN absent a CPG-oriented product. We use non-endemic data to understand our core segmentations to design targeted, on-brand creative that aligns with our client’s primary customers. (Google Analytics offers insights into user behavior, but these insights are primarily based on cookie-laden search activities rather than purchase history. Relying on data that captures converted intent is more dependable for decision-making purposes.)
#MarketingStrategyTips