Always Triangulate Data

Always triangulate data
Always triangulate data

Triangulate your data with Facebook Ads, Analytics and your backend. It has taken me an unimaginable amount of time and effort to reach the level of leaving the insanity behind and scaling results and budgets with sanity. Anyone who is seasoned with Conversion optimization, Lead Generation, Analytics, Google Ads and Facebook Ads Managers knows what I mean so keep your sanity and read this 😱

Triangulated data has a sweet but short-lasting flavor

I have spent a lot of years with testing and experimenting with curiosity surrounding how you can structure your marketing data and use that information for better and smarter strategies and tactics. All these wins and losses are in most cases holding a sweet but short-lasting flavor as our marketing landscape is always changing. These platforms and algorithms are changing so rapidly that I think a lot of marketers are having micro heart attacks when watching performance numbers, do you recognize anyone you know? 

With that said I hope I can share some thoughts that will help anyone avoid mistakes and scale faster regardless if your work with B2B lead generation, B2C, run your eCommerce business or help other agencies as a performance agency. 

Triangulate data with Facebook Ads

What makes or breaks a campaign is something I think you have to reflect around – are you focusing on; 

  • Platform technicalities 
  • Bidding
  • Scaling strategies
  • CBO
  • Hidden targeting options 
  • Formats

Or other stuff that makes a media buying analytical geek go nuts and all this could make you think it’s technical areas that make or breaks your Facebook campaigns. Now reflect on the actual framework your marketing is using and where your activities stand within that framework. A framework could be Growth metholodgy,  Inbound metholodgy, See Think Feel Do Care with Google or maybe even a custom made.

Triangulate data and focus on what is important

Brand

This is a performance factor that not many (from my experience) see as a popular subject among online marketers, but the levels of brand awareness can determine the success of the campaigns in how well known the brand already is. Brand building helps acquisition strategies. (Ever heard about Peter Fields and Les Binet studies?) 

Product or service

How unique, how useful (think utility), emotional and visual appeal (are you talking to my brain, stomach, sex appeal), seasonal or evergreen nature (useful sometimes or always). There is a high probability that you are far away from this, but If you manage larger budgets then your learnings are larger and therefore you are more likely to share those learnings and feedback with product/-sales teams. 

Your value, the offering

How attractive is the offering compared to it’s perceived value. 

Testing

Regardless of what you test in your marketing, tracking and analytics you should test 2-4 weeks (depending on your buying cycle). If you test a shorter timeframe or period you can miss important data – like every second thursday in the month have X. If you go longer with 4-6 weeks you have to much missing cookies – if cookies is your source of data.

Creative IS KING

With a good product or service, you can screw up the targeting, optimization, and bidding, but if you’ve got good creative, you’ll be fine in a lot of cases (but I never do muhaha). This is why I believe that the creative team should never be on short supply when you drive your Formula 1 on the track, you need someone in the intercom helping you adjust things. That team needs to be creative experts with good ideas that stand out, stop the audiences from scrolling an engaging in your creative, and evoke desire. I don’t know about you but I don’t like to compare elephants with elephants so I test different creative angles. I try to keep multiple working angles in as many campaigns setups as possible to take advantage of micro-targeting as in any local, regional or global market there are always people who will respond better to different approaches (remember Cambridge Analytica?).

Funnel, User flows and User Experience

I hope you know the platforms always chase the relevance for the users so running traffic to a broken funnel is a waste of money. Always try to have conversion rate optimization skills and UX science close by and control before, during and after it because the results are very dependent on the on-site experience. Try to collaborate with data from Hotjar polls, heat maps, recordings, constant user surveys, monitor ad feedback, comments, and reviews of your competitors’ movements and run A/B tests on the website. Now we can impact conversion rates, Average Order Values, and Returning Customer Rates and make sure our advertising strategy is always perfectly aligned with the funnel. This translates for me into better results with greater ROAS, greater customer satisfaction fro the business. It’s not free, it comes with a higher level of complexity in the marketing services, but it’s more rewarding with a deeper level of work that makes a business impact over a longer period. 

Targeting

Start simple, scale your work when your confidant. 

Things change

Everyone can dream that they are a mature growth team but the reality is something else. I always try to structure my work depending on the applied marketing framework and I keep an open mind when I try to scale results. The silver lining and silver bullet are out there somewhere but the probability of finding one comes from structure and luck and most people don’t apply structure to the work. You don’t know when a lookalike, custom audience, broad interests, open targeting, auto or manual bidding will work. My recommendation would be to try to test it for yourself because THINGS CHANGE. CHANGE IS AWESOME!

Triangulate data and look for full-funnel ROI and don’t get fixated on Cost of acquisition 

This could sound like a pretty basic thing, but it’s important. When an objective becomes harder to scale it’s depending on the Facebook algorithm to find ready-to-act, ready-to-buy audiences, which means that the Costs Per Acquisition increases. You should always try to work on your creatives and technical tasks, but remember that Upsells, cross-sells, behavior-based sales, Lifetime Value, Returning Customer Rate, and Referrals should get almost ALL your attention. A lot of businesses are trying to build a subscription model into the product or service to exponentially and predictably increase the revenue every month. I don’t think you should let increasing CPA on the front end stop you from scaling your profits. Especially not for B2B lead generation because of the “wiggle room” in order values etc. 

The halo effect 

I would argue that holistic and measurable campaign performance is a myth, but a nice concept to strive for. The days of Mad Men are gone, we know that, and the complete lack of accountability for the returns on advertising spend is also gone in my book. Many business owners have to big zoom installed in their eyewear (mindset) and are to narrow when looking at their dashboards and are convinced they’re seeing the whole holistic picture. This is a common mistake, but the smartest people I know that manage to scale businesses understand that as the ad budget scales, there is always some halo effect of additional sales due to buyers using:

  • Ad blocks
  • Referrals
  • Cross-device conversions
  • Increased brand awareness and more. 

Always triangulate data between Facebook Ads Manager, Analytics and your backend/-store backend and embrace some degree of uncertainty. This is the balance between insanity and sanity for me =)

Triangulate data and value activities for long-term potential

Are you, the team, the strategy or tactic focused on how you spend your marketing budget rather than on how marketing could help drive long-term business objectives. Thinking like this holds businesses back and marketing should be valued for its long-term potential, rather than its short-term efficiencies. Would you agree or disagree?

The new formats and creatives is not the silver bullet 

A common mistake that I have been reflecting about is that many try to follow the new formats and creatives as a marketing solution to revenue problems. They can be great for some cases, but I have seen: 

Stories

Stories can work with a strong lead front-end product or service. Never assume anything in terms of targeting audiences with this. Test, test, and test. 

Be agnostic

Be agnostic about creative styles and test it all with dynamic ad creator: images, carousels, video collections, stories, canvas, lead ads, etc. Don’t assume what will be the winner. 

Facebook expectations

Facebook thinks that you expect to get better results from video, and when a user is watching your video, they could be scrolling through even 10 image ads/-posts, so they will charge you more for impressions resulting in higher CPMs. Have you seen images getting better results than video? 

Videos and dynamic ads is an article in itself, but not everything is effective and you should try to use dynamic ads and personalization and automation as much as you can.

Trust the algorithm

Facebook becomes smarter and smarter. The new Campaign Budget Optimisation feature is working for some and not for others, for testing, scaling or remarketing (maybe to a smaller extent to be risk aware). 

The egos of some ad buyers and marketers are affected because AI is getting better than any human in making decisions. We have to learn to work with it and give it more control by doing broader audiences, longer decision times, larger budgets, more auto placements. And then focus on the creative, on-site experience and LTV. 

End-to-end analytics and attribution

If you like to explore how you can structure you marketing data, please do the homework first and put down your revenue goals, user journey, user flow, full-funnel and the activities you do and the data these aggregate together with your KPI’s. This will help you get going with mapping out your data and find opportunities and if you need some interesting reading;

https://supermetrics.com/

https://www.owox.com/

https://www.kaushik.net/avinash/multi-channel-attribution-modeling-good-bad-ugly-models/

https://growthhackers.com/tags/analytics

The question is – did you find this valuable? I keep on wondering what I take for granted, what I’m assuming, what I’m not asking and not testing etc. 

Have anything on your mind? It’s starts with a friendly call!

I’m glad your curious!

search previous next tag category expand menu location phone mail time cart zoom edit close