Showing Value In 2014: SEO Attribution Deep Dive
Carefully thought-out campaign attribution can show your SEO campaign’s true value to your business. Settling on a model which paints the least-wrong picture is your starting point, but pretty quickly you can make your model sing, providing real value and efficiency to your marketing budget blend.
Recently, I was asked to help a business answer a question: how can I make £5 million of marketing spend deliver as much value as £15 million had previously?
For lots of agencies, this will be a common question type. Companies with bloated post-recession marketing budgets are trying to piece back together cross-channel marketing budgets to gain an edge over their competition and their mantra this time ’round is efficiency.
Obvious questions arose: which channel delivered the most revenue? How do you measure ROI right now? And, my favourite: are your analytics accurate?
Pretty quickly, I was able to establish that they were keen to rewrite their approach across the board as they had a good understanding of the efficiency available from smart online campaigns across the marketing mix. They were insistent that TV would remain a chunk of that mix due to their market. (Sadly, I’ll have to leave the exact market obscure but feel free to assume a highly competitive market with a long history in online marketing. Also: it’s not retail.)
Oh, they also need to consider 7-8 key country targets in a mix of languages.
This seemed like an opportunity to build from the ground up the ideal marketing mix with all facets of the marketing budget allocated to deliver the biggest combined bang for the buck.
Building An Efficient Marketing Mix
There are a couple of big-button areas where the mix priority will be obvious in this case:
• Multinational (Technical) SEO. This is still a huge opportunity area. I’m amazed at that, but delighted for my clients as they get to reap the rewards of lax competition. This is a one-off piece, so it’s not a huge chunk of the overall budget.
• Soft Competition Countries. With a little benchmarking of PPC and SEO competition in each target territory, we’ll get a good idea of the global low-hanging fruit for both of these high-value monthly cost channels. We can also estimate the size of the prize in each case. We’ve built a tool in-house to estimate organic competition levels, and with a little thought, I suspect you can think of a way to measure this, too. (Tip: marketshare, competing pages, rank, dominance of domains across related terms, backlink strength.)
• Mobile Device Competition Gap. We know that the traffic levels for mobile devices are historically higher than their PPC bid spends suggest they should be — this leaves a handy gap that can be usefully exploited (see image below) by clever mobile and tablet segmentation in your PPC budget bid strategy.
• Dynamic Remarketing. Triggered against tightly defined goals, remarketing can be a very efficient use of budget to capture lost conversions. The key is to match precisely relevant information within the ad — so avoid static ads (please), and use inventory to show items that relate to the lost conversion’s likely intent. (While my client isn’t a retail site, they do have “inventory.”)
• Affiliates. When you’re looking at tightly cut back budgets, affiliate marketing is often the first channel to get the chop. Before you consider that, though, let’s get our attribution on. I recommend getting a strong cup of coffee and setting aside a solid day to work through each example given by Avinash in this superb post on the basics behind attribution models. Try out a few examples and see how they would influence your decision, specifically, on cutting back a channel that’s currently driving revenue.
• Email. Similar to the use case for dynamic remarketing, cutting back on email channel budget is often a budgeting failure as this channel’s ability to soak up non-converters as a soft conversion (registration capture, restock availability, new releases/early peek, general info, deals, advocacy/benefits, etc.) and convert them later is unparalleled.
That said, making it work efficiently is often a challenge. Looking at custom time-delay attribution models for each of your site goals that incorporate email as part of the conversion path will give you an idea of current value. It’s likely this value is way more than budget against this activity and so there’s little scope for rebudgeting opportunity here.
• Social. This channel will be looped into SEO promotion budgets in this case, as the business handles this internally, and excellently, at the moment and doesn’t need to account budget against this channel in their mix (lucky them!).
• Offline Brought Online. Using Universal Analytics, as well as call tracking and tagged campaigns everywhere (via URL shorteners for social engagement richness), you should be able to better attribute offline impact on conversion path. Again, crack out a tested, custom attribution model for this. Apply to existing data if you have foresight, wait until the data flows in from your first tagged offline campaign if not.
Your remaining big ticket item to consider is the large monthly budget balance between PPC spend, offline and SEO execution (whether agency or in-house time). Again, modelling for each of these channels will tell you the risk you are incurring in cutting budget. I strongly urge using at least a time-weighted attribution model to inject a little sophistication into your decision.
In the case of this specific query, what was my outcome?
The size of the technical SEO prize and the multinational aspect suggests the biggest returns will come from weighting efficient mid-tail SEO and PPC campaigns heavily across the top three countries. With the platform as a whole benefitting from the technical quick wins, general SEO performance will improve dramatically, and landing pages will deliver better Quality Scores for efficient supporting PPC campaigns across the board.
A small test run offline campaign in each target territory, correctly tagged, will indicate the final size of budget allowed to remain in offline. If the tests fail a set ROI target using our best-guess attribution model, then they get funnelled into late-stage campaign PPC budgets with display and mobile PPC activity being ramped up in the biggest target market (the US).
Affiliate marketing, sadly, does not have a bright future in this mix. In this case, an overly generous affiliate deal due to a saturation of competitors in the market, backed by Google’s heavy clamp down on affiliates’ ability to take large inventory and prosper in SEO, has killed ability to drive cheaper traffic, so a lower value deal likely won’t get many takers.
Reviewing performance after the first offline ad would set the full year marketing blend mix. Performance against some simple KPIs can trigger extra budget and a slow climb back the pre-recession days with a much healthier, underlying mix and a more competitive offering.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.