The concept of performance based SEO campaigns is simple. Formerly known as ‘Pay for Rankings’ campaigns, they’re based upon the agency being paid only as and when they deliver results (or a substantial amount of monthly fees withheld until results are produced). Awesome! If agencies are so confident in their SEO abilities, surely they would be happy to put their money where their mouths are?
On the surface of it, yes. It looks great and believe me, as an agency, if we could make it work it would make our lives a hell of a lot easier when pitching to clients! However, dig below this fragile facade and it can quickly transform into unforeseen traps, potential subterfuge, high levels of complexity and misunderstandings for both the client and the agency alike. We’ve actually spent a considerable amount of time investigating options with some meaty campaigns and the conclusion is often very similar: performance based SEO retainers are very difficult for both the agency and the client.
If you are some sort of savant and have managed to simplify this da Vinci code then come out and claim your glory. Agencies and businesses alike will clamour for your knowledge!
SEO has a dirty and not so distant past. The industry as a whole pays in one way or another for the transgressions of those who have ventured (and still venture) to the side of Black Hat SEO (spam), or simply fail to deliver the promised work. Unfortunately there are many that fall into this category and as a direct result, almost every potential client you speak to has either been directly burnt by an SEO or knows someone who has engaged with an underperforming or spammy campaign.
(We’re also not helped by a baffling amount of terminology, often wishy washy ROI and attribution models, and a service that requires months and months before seeing any significant returns.)
I’ve bullet pointed two major pieces of rhetoric that raise customer’s eyebrows and have therefore naturally contributed to the desire for this concept. I’ve also input how customers that have previously had bad experiences may interpret these statements (in a polarised fashion):
An SEO campaign is a long term commitment
“The client has to pay monthly so it means that as an agency we will get paid for many months, regardless of results”
We can’t guarantee results, we’re not Google
“This is our get out of jail free card. We’ll do our best and that’s what counts, right?”
Now we agree that paying customers are entitled to demand some balancing of these scales. If they’re being asked to jump off a cliff, why shouldn’t the agency tandem free fall with them? It seems reasonable. And after all, you don’t expect to pay someone to build you a house, to then not have a house at the end of it.
The unfortunate fact is that it ain’t quite that simple.
Final word before we dive into the details of why, more often than not, performance based SEO doesn’t work: as the owner of an SEO agency, I would of course want to ensure that I unravel the efficacy of these types of campaigns. And I do. But as I said earlier, if they worked they might actually make my job easier. It would mean that the agencies that delivered great work would win clients at the drop of a hat, but alas, it is not to be and here’s why.
THE DATA ISN’T CLEAR CUT
Let’s get right into it. Whilst we have also listed reasons why these campaigns can be problematic specifically for the agency, and then specifically for the customer, the biggest reason they rarely work is that it’s fiendishly difficult to make the data work.
Rankings as a Metric
As a customer you want to pay when you get results. So what constitutes results? Back in the day, it used to be as simple as rankings (hence ‘Pay for Rankings’) but we all know that your position within the SERPs is not the be all and end all of an SEO campaign. Furthermore, regardless of how indicative rankings are to the success of the campaign, how are you going to measure those rankings? The SERPs change daily between devices, location and preferences…to name a few factors. Which record will be used, do you settle on using a particular tracking tool and agree that the inaccuracies are acceptable? What about local pack?
Finally, in some industries the terminology can change, ebb and flow. As such, through the course of a campaign the primary search terms that you target may change. For obvious reasons this then draws into question the original agreement.
If success is to be objective and data driven, this is a problem for performance based payment terms.
Organic Search Traffic as a Metric
Right. Okay, organic search traffic then.
This seems to be a better metric for success. Figure out what represents a reasonable level of organic search traffic on which to base an expected increase. Set target. Simple. You might even want to set certain ratchets or tiers where portions of fees are due, when certain percentages of targets have been hit (a bit like sales commission). As we said earlier. On the surface this looks like a nice clean way to do things. There are however a number of factors that tend to throw a spanner in the works:
– Not all traffic is equal
This doesn’t take into account traffic quality and this is critical. More traffic does not necessarily mean more sales or conversions. Using organic search traffic as your sole metric means that the financial side of the arrangement is based on something that does not have a one to one correlation with the financials of the business.
– Alternate Influencers
Organic search traffic is not only influenced by your rankings for target keywords. People also use branded search terms to get to your website (i.e by Googling your brand name). They also might use what are commonly known as ‘Navigational’ search terms – i.e they have seen your brand and are looking for it. These might be contextual or descriptive. Whatever the case, this is traffic that can be increased by factors outside of an SEO’s control.
Think social media campaigns, email send outs, billboards, tv adverts. A piece of press or something going viral. Maybe even mistaken identity!
On one hand, is it fair that a customer would then have to pay an agency for increases in traffic that might not be attributable to the agency’s work? If it isn’t fair, how do you go about ensuring that this doesn’t occur? Freeze all other marketing channels, content output, product releases or relations with press for the duration of the campaign? This doesn’t even take into account the problems associated with organic traffic and other channels being attributed to the ‘Direct’ channel via Google Analytics.
– Seasonal Variations
Some businesses operate in industries with significant seasonal variations. A great example of this is the Ski industry. In Winter, business is good. Summer – not so much.
These seasonal variations can add a further layer of complexity to the agreement. What if the campaign finishes prior to the start of the season? Are you basing the increase on a yearly, quarterly or monthly average? Are performance based fees paid out on a monthly or quarterly basis? Admittedly these are probably issues that can be ironed out – but you’re going to want to be crystal clear on all of the above. Que headache.
Conversions as a Metric
As Stoney de Geyter commented in an article for Marketing Land the previous two factors are easy to discount. Paying upon money in the bank again seems to be the only logical step.
Yes, it can work. As long as all of the following factors are perfect and the list of necessary compromises for the agency and client are acceptable to both parties…
– Alternate Influencers
Back again. Sales via organic search can be influenced by other campaigns, so the agency might profit from this work.
Analytics platforms are continually improving in their ability to attribute genuine sales to their relevant marketing channel. This does require that the customer has had this tracking set up for a period of time prior to the campaign commencing. This way, both parties are able to set fair targets. The percentage of businesses in this position is very low indeed, especially amongst those with limited budgets or small internal marketing teams.
In addition, cross-device channel attribution is nigh on impossible to get 100% correct. As an example, if someone uses a transactional search term to find a site but does not then make a purchase. They may subsequently use another device and navigate directly to the website and make a purchase, this conversion would have been instigated by SEO but would be attributed to the ‘Direct’ channel.
Finally, a significant hindrance is the amount of search query data that falls under ‘not provided’ in Google Analytics. Is the performance incentive based purely on transactional keywords rather than branded search? If it is, you need read no further.
– Pre-existing trends
The client may already have a general upward trend in their organic traffic and conversions via organic search. As such, the agency and the client will have to agree a level at which this trend will continue. Clearly this could result in the agency taking credit for a trend changing to their benefit, or an agency losing out when a trend no longer fulfils predictions.
Hopefully the above gives you some insight into why having both parties agree to a performance based remuneration structure that is fair and reasonable for both is incredibly hard when purely on a numbers basis.
TIME & COMPROMISES
Hopefully by now you can see that there are a lot of moving parts in this type of agreement (and we’re only halfway through discussing the topic!). These contracts are not easy to put together and require two key items:
A significant amount of time (and costs) to negotiate, agree and translate to a contract. Your work is not done there – you will have ongoing admin costs over and above a normal SEO retainer to ensure that KPIs are not only hit but documented. This cost is absorbed by both the client and the agency so the retainer needs to be of a certain value to justify this additional cost.
Both parties must be willing to make compromises during negotiations in order for this agreement to be rubber stamped. The likelihood is that one (or both) parties will accept compromises that they otherwise wouldn’t have deemed reasonable, this is obviously a matter of preference and will vary dramatically between different agencies and clients alike. I guess it’s something to be aware of – at what point do these concessions become detrimental to one, or both, parties?
CLIENT SPECIFIC ISSUES
The Pressure is on
The moment that contract is signed, the pressure is on for the agency.
“That’s sort of the point”, I hear you say.
The issue is that for the agency, money is being spent with no (or restricted) money coming in. If you do a bit of reading around the subject, in the past many agencies have used spammy link networks to get quick results in order to collect their pay day. Whilst this tactic is a number of years old now and doesn’t work nearly as well as it might have pre-Penguin, you’re running the risk of engaging with such an unscrupulous agency.
Let’s say that you’ve engaged with an SEO agency that starts with the best intentions. However, eight months pass and they still haven’t been paid. There may become a temptation to start to bend the rules in order to get some quicker wins… rule bending that could have big consequences further down the line. We’re not saying that good agencies would suddenly turn bad, but stranger things have happened in pressure situations.
What if you don’t see the value?
It depends on what metric you’re using to measure success. However, take traffic as an example. Getting an increase in organic search traffic to a website is relatively easy, but how targeted is that traffic?
Agencies don’t even have to be malicious in their targeting of traffic. They could be targeting individuals higher up in the sales funnel with the intention of creating content that subsequently helps convert this traffic into customers. But what if this content is spectacularly successful in terms of traffic, but not in terms of sales? As a client, you can end up paying for something that doesn’t represent the originally intended value.
If it doesn’t work, there’s no risk!
The majority of articles on the web focus on this aspect. Google the subject and it will become apparent that there’s an association between performance based campaigns and the world of spam. The fact that there’s an association with Black Hat SEO techniques should be a sufficient deterrent in itself.
But if it doesn’t work, what’s the risk?
It’s not just the time you’ve lost where you could have been building a really healthy and longer term campaign. In fact that’s the least of your worries. It’s the potential long term effects of any spammy work conducted during the campaign. Onsite spam is relatively easy to rid yourself of, but linkspam can be a real headache. Get caught by any one of Google’s algorithm updates (think Penguin) and this headache morphs instantly into a nightmare.
The long and the short of it is that penalties can haunt websites for months, years and sometimes forever. That’s a rather large risk to take.
The bonuses might be eye watering
I personally don’t think that any good agency would entertain a completely bonus based structure (i.e zero monthly fee). SEO is simply too much hard work and time invested to take that risk. However, they may well entertain a hybrid of sorts.
So the likelihood is that you’ll discuss some sort of baseline retainer in order to cover a portion of costs. It makes it more palatable for the agency, but they’re still accepting more risk than usual.
They’re therefore going to want any on target earnings (OTE) to be in excess of what a normal retainer would have yielded. If they are to adopt a risk, there should be a reward. This may be speculation as I haven’t looked into this in granular detail with other agencies, but if I was to adopt such a risk, the rewards would likely be uncapped. And the OTE would also be substantially more than a normal retainer would have been. Otherwise, why take on the risk? (See football analogy above.)
The customer therefore risks paying large bonuses for performance. As above, performance that might not necessarily represent complete value, depending on the metric.
The Good Guys Don’t Need It
Pre-warning. This isn’t some sort of flex.
The reality is that good agencies don’t necessarily need to accept these types of campaigns. They have the track history, expertise, credentials and references needed for people to buy their services on a normal, monthly retainer. Christiano Ronaldo or Lionel Messi wouldn’t play for a whole season for free, with the possibility of being paid upon winning the Champions League. There are simply too many other clubs offering to pay normally.
As such, on this basis you’re probably not engaging with the very best agency you can. You’re engaging with an agency that needs to take a risk. Which is inherently risky in itself.
AGENCY SPECIFIC ISSUES
SEO is Long Term, with Real Costs.
SEO is not PPC (Google Ads). It’s not a ‘pay to play’ channel.
What do I mean by that?
With Google Ads you can pump money into the platform and as long as you’re paying the right amount per click you can gain instant visibility. You can’t do this with SEO. A good campaign IS a long term commitment and it can produce fantastic results, but it can take multiple months.
A good campaign requires good people, and good people cost money. Put simply, if your agency is based upon pay per performance campaigns and you accept that campaigns can take six, nine or even twelve months to deliver results (for which you get paid), you have to have one hell of a credit line to be able to fund your business for this interim period.
If you do then great, introduce me to your backers. They’ve got some stones on them.
Admittedly you probably won’t be engaging in performance based retainers for every client, that would be particularly brash. But seriously, the risk profile for a business that is built upon doing months and months of free work is going to be particularly high. Sure some may charge a set up fee, some may even charge a smaller retainer but this still means that they’re taking on a large amount of costs for a considerable amount of time with little in return.
Sometimes it needs longer
It’s true. This clearly doesn’t help us sell more SEO campaigns but sometimes campaigns need longer than originally estimated. Competitors may have stepped up their efforts or the agency may have underestimated the requirements. As a rule of thumb, the better the agency, the rarer this will be as a scenario.
However, whilst attempting to balance the scales between customer and agency, the risk liability for the agency has suddenly expanded dramatically. If your agreement is for a set period (more on this later), this could mean that an agency could work for an entire year and not get paid whilst the customer just needs to put some finishing touches to reap all of the rewards. That doesn’t seem all that fair for the agency.
Baseline Metrics are Warped
In order for agencies to be paid for the results they gain for clients, they’re going to need a solid baseline from which they can set goals. In addition, in an ideal world the customer will have had detailed tracking implemented during this time period so that the baseline data is as accurate as possible.
This opens the discussion up to a debate around how the data is interpreted and what constitutes a baseline performance. Whilst this is likely to be the easier conversation to have, the harder conversation is clearly the efficacy of this data.
The Agency Does Not Have Full Control
Aside from the financial and time related risk associated with ‘pay for results’ style campaigns, the second largest issue for agencies is that the SEO results will depend on a lot of factors out of their direct control.
SEO campaigns are complex beasts and it takes the implementation of hundreds of items/factors to get the very best results. On the technical side, what if key onsite optimisation is restricted because of the CMS that the website utilises? This would have an adverse effect on the agency’s ability to earn their fee.
Content creation is a critical aspect of any campaign and can often cause friction between the agency and the client. As an example, the agency might want to be focussed on content that’s directly relatable to their inbound sales funnel and in support of the SEO campaign whereas the customer may want to pursue more editorial, community led content. Clearly the agency would prefer to have full control and autonomy in terms of the content they produce, which is rarely the case.
Finally the data challenges (mentioned earlier in the article) apply both to measuring the success of an SEO campaign and attributing SEO actions to that success. SEO campaigns operate within a multifactorial environment and as such, their results are influenced by multiple factors making it hard to identify the direct cause of each result.
Some of these external factors can be controlled through SEO work, others can’t. There are non campaign related factors that the agency cannot control. Factors that will almost certainly have an influence on the agreement, especially if sales are used as a metric:
- The agency cannot control how products/services are sold
- The agency doesn’t dictate the price point at which products/services are sold and how competitive they are in the market.
- What if bad press or negative reviews decrease sales volumes?
- Perhaps the agency identify that key design changes need to be implemented on the website in order for the campaign to be a success, but the client doesn’t have the budget to make these amends or does not agree with them.
SO DOES PAY-FOR-PERFORMANCE SEO WORK?
Full disclosure – When I started writing this article, I was pretty opposed to the concept of performance based retainers. Hence why the title is as it is.
Time has passed since I started writing and whilst I’m still convinced that for the majority of situations they produce too many question marks for them to be viable for both parties, Yellowball have actually just engaged in a campaign with a small performance aspect so we’ll report back on our experience!
A short conclusion to this article is that performance based retainers are complex and require great tracking, open negotiations and clear objectives. For them to be truly beneficial for both the client and the agency there will need to be some sort of aligning of the stars. They’re not impossible and hopefully this article will provide an initial blueprint for any of you having such discussions, but be aware of the work required to even get to the start of the campaign.
SPREAD THE WORD:
READ MORE RAMBLINGS
How useful are the Yoast Plugin SEO & Readability Scores?
The fabled traffic light system. We look at how Yoast Plugin’s SEO and readability scores can be used and the common misconceptions around trying to chase the greenMay 3rd, 2019
Digital Marketing • SEO
Moz’s Domain Authority is changing – what does this mean?
Moz are rolling out their new Domain Authority, so we’re taking a look at what change is in store, what this means and how it will affect us, if at all.March 5th, 2019
Digital Marketing • SEO
EU fine Google record $5 billion : reaffirms focus on mobile search
This week saw Google receive a record fine of or €4.34 billion, or $5 billion, from the EU commission due to violation of anti-trust (*read ‘monopoly’) laws. Headed upJuly 20th, 2018